Publikationsliste2

Publikationer

2018
  PDF

Stress response and cognitive performance modulation in classroom versus natural environments: A quasi-experimental pilot study with children

År: 2018

Stress response and cognitive performance modulation in classroom versus natural environments: A quasi-experimental pilot study with children

Mygind, L., Stevenson, M. P., Liebst, L. S., Konvalinka, I. & Bentsen, P. 1 Jun 2018 In : International Journal of Environmental Research and Public Health. 15, 6, 15 p., 1098

Publication: Research - peer-reviewJournal article – Annual report year: 2018

Stress during childhood can have mental and somatic health influences that track throughout life. Previous research attributes stress-reducing effects to natural environments, but has mainly focused on adults and often following leisurely relaxation in natural environments. This pilot study explores the impact of natural environments on stress response during rest and mental load and cognitive performance in 47 children aged 10–12 years in a school context. Heart rate variability measures indexing tonic, event, and phasic vagal tone and attention scores were compared across classroom and natural environments. Tonic vagal tone was higher in the natural environment than the classrooms, but no differences were found in event or phasic vagal tone or cognitive performance measures. These findings suggest a situational aspect of the conditions under which natural environments may give rise to stress-buffering influences. Further research is warranted to understand the potential benefits in a real-life context, in particular with respect to the underpinning mechanisms and effects of accumulated exposure over time in settings where children spend large proportions of time in natural environments.

Original languageEnglish
Article number1098
JournalInternational Journal of Environmental Research and Public Health
Volume15
Issue number6
Number of pages15
ISSN1661-7827
DOIs
StatePublished - 1 Jun 2018

  PDF

Phone-based metric as a predictor for basic personality traits

År: 2018

Phone-based metric as a predictor for basic personality traits

Mønsted, B., Mollgaard, A. & Mathiesen, J. 1 Jun 2018 In : Journal of Research in Personality. 74, p. 16-22

Publication: Research - peer-reviewJournal article – Annual report year: 2018

Basic personality traits are believed to be expressed in, and predictable from, smart phone data. We investigate the extent of this predictability using data (n = 636) from the Copenhagen Network Study, which to our knowledge is the most extensive study concerning smartphone usage and personality traits. Based on phone usage patterns, earlier studies have reported surprisingly high predictability of all Big Five personality traits. We predict personality trait tertiles (low, medum, high) from a set of behavioral variables extracted from the data, and find that only extraversion can be predicted significantly better (35.6%) than by a null model. Finally, we show that the higher predictabilities in the literature are likely due to overfitting on small datasets.

Original languageEnglish
JournalJournal of Research in Personality
Volume74
Pages (from-to)16-22
ISSN0092-6566
DOIs
StatePublished - 1 Jun 2018

 

Efficient computation for Bayesian comparison of two proportions

År: 2018

Efficient computation for Bayesian comparison of two proportions

Schmidt, M. N. & Mørup, M. 2018 In : Statistics and Probability Letters. 6 p.

Publication: Research - peer-reviewJournal article – Annual report year: 2018

In Bayesian comparison of two proportions, the exact computation of the evidence involves evaluating a generalized hypergeometric function. Several agreeing, but not identical, expressions for the evidence have been derived in the literature; however, their practical computation (by summing the truncated hypergeometric series) can be troubled by slow convergence or catastrophic cancellation. Using a set of equivalence relations for the generalized hypergeometric function, we derive ten equivalent expressions for the evidence: We show that one of these formulations, which has not previously been studied, is superior in terms of its computational properties. We recommend that this be used instead of existing formulations, and provide an efficient software implementation.

Original languageEnglish
JournalStatistics and Probability Letters
Number of pages6
ISSN0167-7152
DOIs
StateAccepted/In press - 2018

 

Modeling Temporal Dynamics in Functional Brain Connectivity

År: 2018

Modeling Temporal Dynamics in Functional Brain Connectivity

Nielsen, S. F. V. 2018 100 p. (DTU Compute PHD-2018, Vol. 484).

Publication: ResearchPh.D. thesis – Annual report year: 2018

This thesis deals with modeling temporal changes in functional brain connectivity derived from functional magnetic resonance imaging (fMRI). These changes, observed in both task and rest settings, have been coined dynamic functional connectivity (dFC), and are often clustered into a discrete set of so-called dFC states. In the five included research papers, we analyse these repeating patterns of connectivity using Bayesian machine learning methods and relate these to cognitive traits and disease status in different resting-state datasets. In dFC state models, we are faced with many parameter choices, which we in this thesis have tackled using a predictive likelihood framework allowing for quantitative model comparison. Furthermore, this can also be used to assess the relative plausibility of a set of candidate models. We applied this framework to the Wishart mixture model, a probabilistic extension of the sliding-window k-means approach used in many dFC studies. Here, we show that the predictive likelihood can be used to quantify the support for dFC given different window lengths. Furthermore, in another paper we show that the predictive likelihood can be used to choose both the number of states and the model structure in a hidden Markov model (HMM) applied to a highly sampled single subject's resting-state fMRI data. Another way to investigate the relevance of dFC models is to relate them to subject specific cognitive traits or disease status. The former was investigated in a large cohort of healthy subjects' resting-state fMRI data and we found almost no association between the temporal characteristics of the dFC models and the higher order cognitive traits. In another paper we investigated different HMMs ability to distinguish between patients with schizophrenia and healthy controls based on resting-state fMRI data. We found that the simplest characterizations using static FC were adequate for the classiffcation task. Our ndings underline the importance of quantitative evaluation of dFC models and furthermore shows that we need better models that can account for subject variability and noise confounds.
Original languageEnglish
Number of pages100
StateSubmitted - 2018
SeriesDTU Compute PHD-2018
Volume484
ISSN0909-3192

 

Monitoring and modelling of behavioural changes using smartphone and wearable sensing

År: 2018

Monitoring and modelling of behavioural changes using smartphone and wearable sensing

Kamronn, S. D. 2018 110 p. (DTU Compute PHD-2018, Vol. 489).

Publication: ResearchPh.D. thesis – Annual report year: 2018

Increase of sedentary behaviour and obesity has been on the rise for a score of years or more, despite public information campaigns and even the incursion of the latest fad, fitness trackers. The latter were by industry heralded as life–changers that by simple mechanisms would change the behaviour of the wearer to be more active and more healthy. Studies have since shown that they may have an initial positive effect on activity levels and reduced weight, but that it quickly falters and people stop using the trackers altogether. A reoccurring observation seem to be a misunderstanding of what drives human motivation and what it takes to change human behaviour with respect to physical activity. Being reminded of your weight or steps taken throughout the day, is for most people but a mere observation, not an intervention. This misunderstanding, or naïvety, probably stems from conclusions that are drawn from data that are too thin to support them. We propose a paradigm that relies on massive amounts of data, pervasively sampled from smartphones. We show that smartphone data is able to estimate plausible intervention effects from a randomized controlled trial, and through higher sampling frequency and additional modalities, is able to break up the estimated effects into contextual pieces that can be used to better understand behavioural aspects. We further show that by using a model that adapts to each individual, we can predict a persons total energy expenditure accurately from the same data. A novel model to recognise human activity semi–supervised and from multiple datasets, is presented. The model combines convolutional neural networks to extract hierarchical features and recurrent neural networks to model temporal dependencies. This is combined with recent developments in domain adaptation where domain separation is penalised through adversarial training of an auxiliary classifier. Lastly a model is presented that fully unsupervised is able to learn latent states that naturally decompose into static and dynamic representations. The static representations are learnt as a function that maps a high–dimensional observation into a low–dimensional code that is dependent on a structured prior distribution that governs the dynamical system
Original languageEnglish
Number of pages110
StateSubmitted - 2018
SeriesDTU Compute PHD-2018
Volume489
ISSN0909-3192

 

Testing a Model of Destination Image Formation: Application of Nonparametric Bayesian Relational Modeling to Destination Image Analysis

År: 2018

Testing a Model of Destination Image Formation: Application of Nonparametric Bayesian Relational Modeling to Destination Image Analysis

Glückstad, F. K., Schmidt, M. N. & Mørup, M. 2018 2018 Global Marketing Conference at Tokyo Proceedings. Choi, J. (ed.). Global Alliance of Marketing and Management Associations, Vol. 2018, p. 63-64 2 p. (Global Fashion Management Conference).

Publication: Research - peer-reviewArticle in proceedings – Annual report year: 2018

This presentation introduces a methodological framework that analyzes a model of destination image formation (Baloglu & McCleary 1999; Beerli & Martin 2004). Specifically, the main aims of this study are to investigate what type of stimulus factors (information sources) are connected to the formation of destination image, and to explore if there is a connection between their strength of willingness to visit a destination and their patterns to associate with the destination. The study employs an advanced nonparametric Bayesian relational model (Glückstad, Herlau, Schmidt, Rzepka, Araki and Mørup 2013; Mørup, Glückstad, Herlau & Schmidt, 2014) for a two-steps analysis . The first step attempts to segment consumers according to patterns of attributes consumers associate with three arbitrary selected destinations. The second step statistically analyzes latent structural patterns per segment by contrasting two independent datasets, one consisting of information sources and members of a segment and another consisting of destination attributes and the members of the segment. The results of two-steps analysis demonstrated that patterns of attributes respondents associate with the three selected destinations differ across individuals and the applied method enabled to segment respondents according to the differences, and consumers’ associations, their willingness to visit the destinations and types of information sources they have accessed to learn about the destinations are connected to each other.
Original languageEnglish
Title of host publication2018 Global Marketing Conference at Tokyo Proceedings
EditorsJeonghye Choi
Number of pages2
Volume2018
PublisherGlobal Alliance of Marketing and Management Associations
Publication date2018
Pages63-64
DOIs
StatePublished - 2018
Event2018 Global Marketing Conference at Tokyo - Tokyo, Japan

Conference

Conference2018 Global Marketing Conference at Tokyo
CountryJapan
CityTokyo
Period26/07/201829/07/2018
SeriesGlobal Fashion Management Conference
ISSN1976-8699

 

Stability and Similarity of Clusters under Reduced Response Data

År: 2018

Stability and Similarity of Clusters under Reduced Response Data

Litong-Palima, M., Albers, K. J. & Kano Glückstad, F. 2018

Publication: Research - peer-reviewPaper – Annual report year: 2018

This study presents a validated recommendation on how to shorten the surveys while still obtaining segmentation-based insights that are consistent with the analysis of the full length version of the same survey. We use latent class analysis to cluster respondents based on their responses to a survey on human values. We first define the clustering performance based on stability and similarity measures for ten random subsamples relative to the complete set. We find foremost that the use of true binary scale can potentially reduce survey completion time while still providing sufficient response information to derive clusters with characteristics that resemble those obtained with the full Likert scale version. The main motivation for this study is to provide a baseline performance of a standard clustering tool for cases when it is preferable or necessary to limit survey scope, in consideration of issues like respondent fatigue or resource constraints.
Original languageEnglish
Publication date2018
Number of pages4
StatePublished - 2018
Event32nd Annual Conference of the Japanese Society for Artificial Intelligence - Kagoshima, Japan

Conference

Conference32nd Annual Conference of the Japanese Society for Artificial Intelligence
CountryJapan
CityKagoshima
Period05/06/201808/06/2018

 

A Deep Learning Approach for Real-Time Detection of Atrial Fibrillation

År: 2018

A Deep Learning Approach for Real-Time Detection of Atrial Fibrillation

Andersen, R. S., Peimankar, A. & Puthusserypady, S. 2018 In : Expert Systems with Applications. 24 p.

Publication: Research - peer-reviewJournal article – Annual report year: 2018

Goal: To develop a robust and real-time approach for automatic detection of Atrial Fibrillation (AF) in long-term electrocardiogram (ECG) recordings using deep learning (DL). Method: An end-to-end model combining the Convolutional- and Recurrent-Neural Networks (CNN and RNN) was proposed to extract high level features from segments of RR intervals (RRIs) in order to classify them as AF or normal sinus rhythm (NSR). Results: The model was trained and validated on three different databases including a total of 89 subjects. It achieved a sensitivity and specificity of 98.98% and 96.95% respectively, validated through a 5-fold cross-validation. Additionally, the proposed model was found to be computationally efficient and it was capable of analyzing 24 hours of ECG recordings in less than one second. The proposed algorithm was also tested on the unseen datasets to examine its robustness in detecting AF for new recordings which resulted in 98.96% and 86.04% for specificity and sensitivity, respectively. Conclusion: Compared to the state-of-the-art models evaluated on standard benchmark ECG datasets, the proposed model produced better performance in detecting AF. Additionally, since the model learns features directly from the data, it avoids the need for clever/cumbersome feature engineering.

Original languageEnglish
JournalExpert Systems with Applications
Number of pages24
ISSN0957-4174
DOIs
StateAccepted/In press - 2018

 

EEG Theta Power Is an Early Marker of Cognitive Decline in Dementia due to Alzheimer's Disease

År: 2018

EEG Theta Power Is an Early Marker of Cognitive Decline in Dementia due to Alzheimer's Disease

Musaeus, C. S., Engedal, K., Høgh, P., Jelic, V., Mørup, M., Naik, M., Oeksengaard, A-R., Snaedal, J., Wahlund, L-O., Waldemar, G. & Andersen, B. B. 2018 In : Journal of Alzheimer's Disease. 64, 4, p. 1359-1371

Publication: Research - peer-reviewJournal article – Annual report year: 2018

Background: Quantitative EEG (qEEG) power could potentially be used as a diagnostic tool for Alzheimer's disease (AD) and may further our understanding of the pathophysiology. However, the early qEEG power changes of AD are not well understood.Objective: To investigate the early changes in qEEG power and the possible correlation with memory function and cerebrospinal fluid biomarkers. In addition, whether qEEG power could discriminate between AD, mild cognitive impairment (MCI), and older healthy controls (HC) at the individual level.Methods: Standard EEGs from 138 HC, 117 MCI, and 117 AD patients were included from six Nordic memory clinics. All EEGs were recorded consecutively before the diagnosis and were not used for the consensus diagnosis. Absolute and relative power was calculated for both eyes closed and open condition.Results: At group level using relative power, we found significant increases globally in the theta band and decreases in high frequency power in the temporal regions for eyes closed for AD and, to a lesser extent, for MCI compared to HC. Relative theta power was significantly correlated with multiple neuropsychological measures and had the largest correlation coefficient with total tau. At the individual level, the classification rate for AD and HC was 72.9% for relative power with eyes closed.Conclusion: Our findings suggest that the increase in relative theta power may be the first change in patients with dementia due to AD. At the individual level, we found a moderate classification rate for AD and HC when using EEGs alone.
Original languageEnglish
JournalJournal of Alzheimer's Disease
Volume64
Issue number4
Pages (from-to)1359-1371
ISSN1387-2877
DOIs
StatePublished - 2018

 

Testing group differences in state transition structure of dynamic functional connectivity models

År: 2018

Testing group differences in state transition structure of dynamic functional connectivity models

Nielsen, S. F. V., Vidaurre, D., Madsen, K. H., Schmidt, M. N. & Mørup, M. 2018 Proceedings of 2018 International Workshop on Pattern Recognition in Neuroimaging. IEEE, p. 1-4

Publication: Research - peer-reviewArticle in proceedings – Annual report year: 2018

Understanding the origins of intrinsic time-varying functional connectivity remains a challenge in the neuroimaging community. However, some associations between dynamic functional connectivity (dFC) and behavioral traits have been observed along with gender differences. We propose a permutation testing framework to investigate dynamic differences between groups of subjects. In particular, we investigate differences in fractional occupancy, state persistency and the full transition probability matrix. We demonstrate our framework on resting state functional magnetic resonance imaging data from 820 healthy young adults from the Human Connectome Project considering two prominent dFC models, namely sliding-window k-means and the Gaussian hidden Markov model. The variables showing consistent significant dynamic differences were limited to gender and the degree of motion in the scanner. We observe for the data considered that a large sample size (here 500 subjects) is needed to to draw reliable conclusions about the significance of those variables. Our results point to dynamic features providing limited information with regard to behavioral traits despite a relatively large sample size.
Original languageEnglish
Title of host publicationProceedings of 2018 International Workshop on Pattern Recognition in Neuroimaging
PublisherIEEE
Publication date2018
Pages1-4
ISBN (print)978-1-5386-4291-7
DOIs
StatePublished - 2018
Event8th International Workshop on Pattern Recognition in Neuroimaging - Singapore, Singapore

Conference

Conference8th International Workshop on Pattern Recognition in Neuroimaging
LocationCentre for Life Sciences at the National University of Singapore
CountrySingapore
CitySingapore
Period12/06/201814/06/2018

 

A gaze interactive assembly instruction with pupillometric recording

År: 2018

A gaze interactive assembly instruction with pupillometric recording

Hansen, J., Mardanbegi, D., Biermann, F. & Bækgaard, P. 2018 In : Behavior Research Methods. 50, 4, p. 1723-1733

Publication: Research - peer-reviewJournal article – Annual report year: 2018

This paper presents a study of a gaze interactive digital assembly instruction that provides concurrent logging of pupil data in a realistic task setting. The instruction allows hands-free gaze dwells as a substitute for finger clicks, and supports image rotation as well as image zooming by head movements. A user study in two LEGO toy stores with 72 children showed it to be immediately usable by 64 of them. Data logging of view-times and pupil dilations was possible for 59 participants. On average, the children spent half of the time attending to the instruction (S.D. 10.9%). The recorded pupil size showed a decrease throughout the building process, except when the child had to back-step: a regression was found to be followed by a pupil dilation. The main contribution of this study is to demonstrate gaze-tracking technology capable of supporting both robust interaction and concurrent, non-intrusive recording of gaze- and pupil data in-the-wild. Previous research has found pupil dilation to be associated with changes in task effort. However, other factors like fatigue, head motion, or ambient light may also have an impact. The final section summarizes our approach to this complexity of real-task pupil data collection and makes suggestions for how future applications may utilize pupil information.
Original languageEnglish
JournalBehavior Research Methods
Volume50
Issue number4
Pages (from-to)1723-1733
ISSN1554-3528
DOIs
StatePublished - 2018

 

A Deep Learning Approach to Identify Local Structures in Atomic-Resolution Transmission Electron Microscopy Images

År: 2018

Recording atomic-resolution transmission electron microscopy (TEM) images isbecoming increasingly routine. A new bottleneck is then analyzing thisinformation, which often involves time-consuming manual structuralidentification. We have developed a deep learning-based algorithm forrecognition of the local structure in TEM images, which is stable to microscopeparameters and noise. The neural network is trained entirely from simulationbut is capable of making reliable predictions on experimental images. We applythe method to single sheets of defected graphene, and to metallic nanoparticleson an oxide support.
Original languageEnglish
Article number1800037
JournalAdvanced Theory and Simulations
Volume1
Issue number8
Number of pages12
ISSN2513-0390
DOIs
StatePublished - 2018

  PDF

Evidence for a Conserved Quantity in Human Mobility

År: 2018

Evidence for a Conserved Quantity in Human Mobility

Alessandretti, L., Sapiezynski, P., Sekara, V., Lehmann, S. & Baronchelli, A. 2018 In : Nature Human Behaviour. 2, p. 485–491

Publication: Research - peer-reviewJournal article – Annual report year: 2018

Recent seminal works on human mobility have shown that individuals constantly exploit a small set of repeatedly visited locations. A concurrent literature has emphasized the explorative nature of human behavior, showing that the number of visited places grows steadily over time. How to reconcile these seemingly contradicting facts remains an open question. Here, we analyze high-resolution multi-year traces of $\sim$40,000 individuals from 4 datasets and show that this tension vanishes when the long-term evolution of mobility patterns is considered. We reveal that mobility patterns evolve significantly yet smoothly, and that the number of familiar locations an individual visits at any point is a conserved quantity with a typical size of $\sim$25 locations. We use this finding to improve state-of-the-art modeling of human mobility. Furthermore, shifting the attention from aggregated quantities to individual behavior, we show that the size of an individual's set of preferred locations correlates with the number of her social interactions. This result suggests a connection between the conserved quantity we identify, which as we show can not be understood purely on the basis of time constraints, and the `Dunbar number' describing a cognitive upper limit to an individual's number of social relations. We anticipate that our work will spark further research linking the study of Human Mobility and the Cognitive and Behavioral Sciences.
Original languageEnglish
JournalNature Human Behaviour
Volume2
Pages (from-to)485–491
ISSN2397-3374
DOIs
StatePublished - 2018

 

The Dangers of Following Trends in Research: Sparsity and Other Examples of Hammers in Search of Nails

År: 2018

The Dangers of Following Trends in Research: Sparsity and Other Examples of Hammers in Search of Nails

Adali, T., Trussell, H. J., Hansen, L. K. & Calhoun, V. D. 2018 In : Proceedings of the IEEE. 106, 6, p. 1014-1018

Publication: Research - peer-reviewJournal article – Annual report year: 2018

Trends, they are not only for the fashion industry after all. Within the engineering and computer science research communities as well, we periodically observe the phenomenon, see how certain methods suddenly start receiving particular attention, and sometimes, even though they emerge as an attractive solution for a given set of problems, they tend to become a hammer looking for new nails. At first, using a new method on old problems is the natural and reasonable way to proceed. There have been remarkable successes achieved through the adoption of a tool from another field or a new way of looking at old problems that brings new insights and solutions.
Original languageEnglish
JournalProceedings of the IEEE
Volume106
Issue number6
Pages (from-to)1014-1018
ISSN0018-9219
DOIs
StatePublished - 2018

 

Understanding Mindsets Across Markets, Internationally: A Public-private Innovation Project for Developing a Tourist Data Analytic Platform

År: 2018

Understanding Mindsets Across Markets, Internationally: A Public-private Innovation Project for Developing a Tourist Data Analytic Platform

Albers, K. J., Schmidt, M. N., Litong-Palima, M., Mørup, M., Bonnevie, R. & Kano Glückstad, F. 2018 Proceedings of the 2018 IEEE 42nd Annual Computer Software and Applications Conference (COMPSAC). S. R., S. I. A., C. D., T. C., W. C., M. N., E. T., S. C., C-H. L., H. T., J-J. Y., T. A., Z. Z. & K. H. (eds.). IEEE, p. 159-164 6 p.

Publication: Research - peer-reviewArticle in proceedings – Annual report year: 2018

This paper presents an ongoing public-private innovation project that integrates unsupervised machine learning tools and a marketing theory, in order to analyze segment-based attitudes and behaviors of tourists. Our case study involving the major governmental tourism stakeholders emphasizes the importance of developing a user-friendly data analytic pipeline that carefully considers users' data collection procedure, easy access to the back-office computation algorithms, an interactive output data analysis workflow and its visualization. At the end of this paper, we present our vision to further develop a cloud-based tourist data collection platform.
Original languageEnglish
Title of host publicationProceedings of the 2018 IEEE 42nd Annual Computer Software and Applications Conference (COMPSAC)
EditorsSorel Reisman , Sheikh Iqbal Ahamed , Claudio Demartini , Thomas Conte , William Claycomb , Motonori Nakamura , Edmundo Tovar , Stelvio Cimato , Chung-Horng Lung , Hiroki Takakura , Ji-Jiang Yang , Toyokazu Akiyama , Zhiyong Zhang , Kamrul Hasan
Number of pages6
PublisherIEEE
Publication date2018
Pages159-164
ISBN (print)978-1-5386-2667-2
ISBN (electronic)978-1-5386-2666-5
DOIs
StatePublished - 2018
Event42nd Ieee Annual Computer Software and Applications Conference - Tokyo, Japan

Conference

Conference42nd Ieee Annual Computer Software and Applications Conference
CountryJapan
CityTokyo
Period23/07/201827/07/2018

 

The body talks: Sensorimotor communication and its brain and kinematic signatures

År: 2018

The body talks: Sensorimotor communication and its brain and kinematic signatures

Pezzulo, G., Donnarumma, F., Dindo, H., D'Ausilio, A., Konvalinka, I. & Castelfranchi, C. 2018 In : Physics of Life Reviews.

Publication: Research - peer-reviewJournal article – Annual report year: 2018

Human communication is a traditional topic of research in many disciplines such as psychology, linguistics and philosophy, all of which mainly focused on language, gestures and deictics. However, these do not constitute the sole channels of communication, especially during online social interaction, where instead an additional critical role may be played by sensorimotor communication (SMC). SMC refers here to (often subtle) communicative signals embedded within pragmatic actions – for example, a soccer player carving his body movements in ways that inform a partner about his intention, or to feint an adversary; or the many ways we offer a glass of wine, rudely or politely. SMC is a natural form of communication that does not require any prior convention or any specific code. It amounts to the continuous and flexible exchange of bodily signals, with or without awareness, to enhance coordination success; and it is versatile, as sensorimotor signals can be embedded within every action. SMC is at the center of recent interest in neuroscience, cognitive psychology, human-robot interaction and experimental semiotics; yet, we still lack a coherent and comprehensive synthesis to account for its multifaceted nature. Some fundamental questions remain open, such as which interactive scenarios promote or not promote SMC, what aspects of social interaction can be properly called communicative and which ones entail a mere transfer of information, and how many forms of SMC exist and what we know (or still don't know) about them from an empirical viewpoint. The present work brings together all these separate strands of research within a unified overarching, multidisciplinary framework for SMC, which combines evidence from kinematic studies of human-human interaction and computational modeling of social exchanges.
Original languageEnglish
JournalPhysics of Life Reviews
ISSN1571-0645
DOIs
StateAccepted/In press - 2018

 

Statistical models for wifi data and educational peer review

År: 2018

Statistical models for wifi data and educational peer review

Wind, D. K. 2018 DTU Compute. 123 p. (DTU Compute PHD-2018, Vol. 482).

Publication: ResearchPh.D. thesis – Annual report year: 2018

With a growing amount of available data, the approach we take in working with and investigating this data is of paramount importance. While the scientific method is underlying for data science as well, a modern approach to solving data-based problems should be more iterative since the issue of having too much data is becoming as common as having too little data. In this work we describe an agile approach to data science called lean data science and give examples of how to approach problems this way. We then describe our work on two specific problems, namely inferring other data sources from WiFi data and effectively scaffolding educational peer review. From The Copenhagen Network Study [Stopczynski et al., 2014] we have been able to work with a dataset collected using more than 1,000 smartphones from students over multiple years. Using this dataset we look at how well WiFi scans are able to replace other data sources such as Bluetooth and GPS. We show that WiFi data can accurately detect so-called stop-locations of the same quality as state-of-the-art GPS-based methods and that WiFi data can mirror Bluetooth data for the purpose of detecting face-to-face interactions between people. Peergrade is a web-based system for facilitating educational peer review built by us. Conceptually the idea of having students review work by other students serves many purposes, including a potential for saving time on grading for teachers and a way to train taxonomically complex skills for students. Because peer review is a complex process (many people reviewing many people) and requires things such as anonymity, evaluation criteria and the ability for instructors to moderate, it is best facilitated using a digital tool. We first describe Peergrade and some of the features it offers to educators. We then describe different research projects around Peergrade that attempts to help quantify the quality of reviews, allocate reviewers in an optimal way and automatically flag problematic feedback for moderation.
Original languageEnglish
PublisherDTU Compute
Number of pages123
StateSubmitted - 2018
SeriesDTU Compute PHD-2018
Volume482
ISSN0909-3192

 

Structured Bayesian Approximate Inference

År: 2018

Structured Bayesian Approximate Inference

Bonnevie, R. 2018 DTU Compute. 146 p. (DTU Compute PHD-2018, Vol. 481).

Publication: ResearchPh.D. thesis – Annual report year: 2018

This thesis seeks to investigate different facets of the class of Bayesian probabilistic models where the random variables exhibit strong dependencies and simultaneously lack any conditional independence structure, preventing the distribution from being factorized. Without a tractable factorization, a lot of standard inference algorithms become unavailable. We consider the application of variational inference from two different perspectives. In the first scenario we start with an extended model with conditional independence structure, and try to take the auxiliary parameters out of the equation in an optimal manner in a process emulating marginalization. In the second scenario, we tackle the variational problem directly, trying to find an efficient way to represent unfactorized models in an efficient manner, by introducing a separate form of structure to ensure efficiency. For discrete models, we find efficient approximations in the tensor literature that can model structure without sacrificing tractability. Finally, we consider a problem involving Gaussian processes that take random variables as input, leading to an inefficient inference problem. We develop a procedure that allows the stochastic component of the random input to be integrated into the kernel of the Gaussian process.
Original languageEnglish
PublisherDTU Compute
Number of pages146
StateSubmitted - 2018
SeriesDTU Compute PHD-2018
Volume481
ISSN0909-3192

 

Probabilistic sparse non-negative matrix factorization

År: 2018

Probabilistic sparse non-negative matrix factorization

Hinrich, J. L. & Mørup, M. 2018 Latent Variable Analysis and Signal Separation: 14th International Conference, LVA/ICA 2018, Guildford, UK, July 2–5, 2018, Proceedings. Y. D., S. G., R. M., M. D. P. & D. W. (eds.). p. 488-498 11 p. (Lecture Notes in Computer Science, Vol. 10891).

Publication: Research - peer-reviewArticle in proceedings – Annual report year: 2018

In this paper, we propose a probabilistic sparse non-negative matrix factorization model that extends a recently proposed variational Bayesian non-negative matrix factorization model to explicitly account for sparsity. We assess the influence of imposing sparsity within a probabilistic framework on either the loading matrix, score matrix, or both and further contrast the influence of imposing an exponential or truncated normal distribution as prior. The probabilistic methods are compared to conventional maximum likelihood based NMF and sparse NMF on three image datasets; (1) A (synthetic) swimmer dataset, (2) The CBCL face dataset, and (3) The MNIST handwritten digits dataset. We find that the probabilistic sparse NMF is able to automatically learn the level of sparsity and find that the existing probabilistic NMF as well as the proposed probabilistic sparse NMF prunes inactive components and thereby automatically learns a suitable number of components. We further find that accounting for sparsity can provide more part based representations but for the probabilistic modeling the choice of priors and how sparsity is imposed can have a strong influence on the extracted representations.
Original languageEnglish
Title of host publicationLatent Variable Analysis and Signal Separation : 14th International Conference, LVA/ICA 2018, Guildford, UK, July 2–5, 2018, Proceedings
EditorsYannick Deville , Sharon Gannot , Russell Mason , Mark D. Plumbley , Dominic Ward
Number of pages11
Publication date2018
Pages488-498
ISBN (print)978-3-319-93763-2
ISBN (electronic)978-3-319-93764-9
DOIs
StatePublished - 2018
Event14th International Conference on Latent Variable Analysis and Signal Separation - Guildford, United Kingdom

Conference

Conference14th International Conference on Latent Variable Analysis and Signal Separation
LocationUniversity of Surrey
CountryUnited Kingdom
CityGuildford
Period02/07/201806/07/2018
SeriesLecture Notes in Computer Science
Volume10891
ISSN0302-9743

 

Complex Spreading Phenomena in Social Systems: Influence and Contagion in Real-World Social Networks

År: 2018

Complex Spreading Phenomena in Social Systems: Influence and Contagion in Real-World Social Networks

Jørgensen, S. L. (ed.) & Ahn, Y-Y. 2018 Springer. 361 p. (Computational Social Sciences ).

Publication: Research - peer-reviewBook – Annual report year: 2018

This text is about spreading of information and influence in complex networks. Although previously considered similar and modeled in parallel approaches, there is now experimental evidence that epidemic and social spreading work in subtly different ways. While previously explored through modeling, there is currently an explosion of work on revealing the mechanisms underlying complex contagion based on big data and data-driven approaches.

This volume consists of four parts. Part 1 is an Introduction, providing an accessible summary of the state-of-the-art. Part 2 provides an overview of the central theoretical developments in the field. Part 3 describes the empirical work on observing spreading processes in real-world networks. Finally, Part 4 goes into detail with recent and exciting new developments: dedicated studies designed to measure specific aspects of the spreading processes, often using randomized control trials to isolate the network effect from confounders, such as homophily.

Each contribution is authored by leading experts in the field. This volume, though based on technical selections of the most important results on complex spreading, remains quite accessible to the newly interested. The main benefit to the reader is that the topics are carefully structured to take the novice to the level of expert on the topic of social spreading processes. This book will be of great importance to a wide field: from researchers in physics, computer science, and sociology to professionals in public policy and public health.
Original languageEnglish
PublisherSpringer
Number of pages361
ISBN (print)978-3-319-77331-5
ISBN (electronic)978-3-319-77332-2
DOIs
StatePublished - 2018
SeriesComputational Social Sciences
ISSN2509-9574

  PDF

Constrained information flows in temporal networks reveal intermittent communities

År: 2018

Constrained information flows in temporal networks reveal intermittent communities

Aslak, U., Rosvall, M. & Lehmann, S. 2018 In : Physical Review E. 97, 6, 10 p., 062312

Publication: Research - peer-reviewJournal article – Annual report year: 2018

Many real-world networks represent dynamic systems with interactions that change over time, often in uncoordinated ways and at irregular intervals. For example, university students connect in intermittent groups that repeatedly form and dissolve based on multiple factors, including their lectures, interests, and friends. Such dynamic systems can be represented as multilayer networkswhere each layer represents a snapshot of the temporal network. In this representation, it is crucial that the links between layers accurately capture real dependencies between those layers. Often, however, these dependencies are unknown. Therefore, current methods connect layers based on simplistic assumptions that do not capture node-level layer dependencies. For example, connecting every node to itself in other layers with the same weight can wipe out dependencies between intermittent groups, making it difficult or even impossible to identify them. In this paper, we present a principled approach to estimating node-level layer dependencies based on the network structure within each layer. We implement our node-level coupling method in the community detection framework Infomap and demonstrate its performance compared to current methods on synthetic and real temporal networks. We show that our approach more effectively constrains information inside multilayer communities so that Infomap can better recover planted groups in multilayer benchmark networks that represent multiple modeswith different groups and better identify intermittent communities in real temporal contact networks. These results suggest that node-level layer coupling can improve the modeling of information spreading in temporal networks and better capture intermittent community structure.
Original languageEnglish
Article number062312
JournalPhysical Review E
Volume97
Issue number6
Number of pages10
ISSN2470-0045
DOIs
StatePublished - 2018

 

Skull segmentation from MR scans using a higher-order shape model based on convolutional restricted Boltzmann machines

År: 2018

Skull segmentation from MR scans using a higher-order shape model based on convolutional restricted Boltzmann machines

Puonti, O., Van Leemput, K., Nielsen, J. D., Bauer, C., Siebner, H. R., Madsen, K. H. & Thielscher, A. 2018 Medical Imaging 2018: Image Processing. E. D. A. & . B. A. L. (eds.). SPIE - International Society for Optical Engineering, Vol. 10574, 8 p. (Proceedings of SPIE, the International Society for Optical Engineering).

Publication: Research - peer-reviewArticle in proceedings – Annual report year: 2018

Transcranial brain stimulation (TBS) techniques such as transcranial magnetic stimulation (TMS), transcranial direct current stimulation (tDCS) and others have seen a strong increase as tools in therapy and research within the last 20 years. In order to precisely target the stimulation, it is important to accurately model the individual head anatomy of a subject. Of particular importance is accurate reconstruction of the skull, as it has the strongest impact on the current pathways due to its low conductivity. Thus providing automated tools, which can reliably reconstruct the anatomy of the human head from magnetic resonance (MR) scans would be highly valuable for the application of transcranial stimulation methods. These head models can also be used to inform source localization methods such as EEG and MEG.Automated segmentation of the skull from MR images is, however, challenging as the skull emits very little signal in MR. In order to avoid topological defects, such as holes in the segmentations, a strong model of the skull shape is needed. In this paper we propose a new shape model for skull segmentation based on the so-called convolutional restricted Boltzmann machines (cRBMs). Compared to traditionally used lower-order shape models, such as pair-wise Markov random fields (MRFs), the cRBMs model local shapes in larger spatial neighborhoods while still allowing for efficient inference. We compare the skull segmentation accuracy of our approach to two previously published methods and show significant improvement.
Original languageEnglish
Title of host publicationMedical Imaging 2018: Image Processing
EditorsElsa D. Angelini , Bennett A. Landman
Number of pages8
Volume10574
PublisherSPIE - International Society for Optical Engineering
Publication date2018
DOIs
StatePublished - 2018
EventSPIE Medical Imaging 2018 - Houston, United States

Conference

ConferenceSPIE Medical Imaging 2018
CountryUnited States
CityHouston
Period10/02/201815/02/2018
SeriesProceedings of SPIE, the International Society for Optical Engineering
ISSN0277-786X

  PDF

Rigorous optimisation of multilinear discriminant analysis with Tucker and PARAFAC structures

År: 2018

Rigorous optimisation of multilinear discriminant analysis with Tucker and PARAFAC structures

Frølich, L., Andersen, T. S. & Mørup, M. 2018 In : B M C Bioinformatics. 19, 1, 15 p., 197197

Publication: Research - peer-reviewJournal article – Annual report year: 2018

Background: We propose rigorously optimised supervised feature extraction methods for multilinear data based on Multilinear Discriminant Analysis (MDA) and demonstrate their usage on Electroencephalography (EEG) and simulated data. While existing MDA methods use heuristic optimisation procedures based on an ambiguous Tucker structure, we propose a rigorous approach via optimisation on the cross-product of Stiefel manifolds. We also introduce MDA methods with the PARAFAC structure. We compare the proposed approaches to existing MDA methods and unsupervised multilinear decompositions.Results: We find that manifold optimisation substantially improves MDA objective functions relative to existing methods and on simulated data in general improve classification performance. However, we find similar classification performance when applied to the electroencephalography data. Furthermore, supervised approaches substantially outperform unsupervised mulitilinear methods whereas methods with the PARAFAC structure perform similarly to those with Tucker structures. Notably, despite applying the MDA procedures to raw Brain-Computer Interface data, their performances are on par with results employing ample pre-processing and they extract discriminatory patterns similar to the brain activity known to be elicited in the investigated EEG paradigms.Conclusion: The proposed usage of manifold optimisation constitutes the first rigorous and monotonous optimisation approach for MDA methods and allows for MDA with the PARAFAC structure. Our results show that MDA methods applied to raw EEG data can extract discriminatory patterns when compared to traditional unsupervised multilinear feature extraction approaches, whereas the proposed PARAFAC structured MDA models provide meaningful patterns of activity.
Original languageEnglish
Article number197197
JournalB M C Bioinformatics
Volume19
Issue number1
Number of pages15
ISSN1471-2105
DOIs
StatePublished - 2018

 

Machine learning-based screening of complex molecules for polymer solar cells

År: 2018

Polymer solar cells admit numerous potential advantages including low energy payback time and scalable high-speed manufacturing, but the power conversion efficiency is currently lower than for their inorganic counterparts. In a Phenyl-C_61-Butyric-Acid-Methyl-Ester (PCBM)-based blended polymer solar cell, the optical gap of the polymer and the energetic alignment of the lowest unoccupied molecular orbital (LUMO) of the polymer and the PCBM are crucial for the device efficiency. Searching for new and better materials for polymer solar cells is a computationally costly affair using density functional theory (DFT) calculations. In this work, we propose a screening procedure using a simple string representation for a promising class of donor-acceptor polymers in conjunction with a grammar variational autoencoder. The model is trained on a dataset of 3989 monomers obtained from DFT calculations and is able to predict LUMO and the lowest optical transition energy for unseen molecules with mean absolute errors of 43 and 74 meV, respectively, without knowledge of the atomic positions. We demonstrate the merit of the model for generating new molecules with the desired LUMO and optical gap energies which increases the chance of finding suitable polymers by more than a factor of five in comparison to the randomised search used in gathering the training set.
Original languageEnglish
Article number241735
JournalJournal of Chemical Physics
Volume148
Number of pages13
ISSN0021-9606
DOIs
StatePublished - 2018

  PDF

A Fitts' law study of click and dwell interaction by gaze, head and mouse with a head-mounted display

År: 2018

A Fitts' law study of click and dwell interaction by gaze, head and mouse with a head-mounted display

Hansen, J. P., Rajanna, V., MacKenzie, I. S. & Bækgaard, P. 2018 Proceedings of the Workshop on Communication by Gaze Interaction . Association for Computing Machinery, p. 1-5 Article No. 7

Publication: Research - peer-reviewArticle in proceedings – Annual report year: 2018

Gaze and head tracking, or pointing, in head-mounted displays enables new input modalities for point-select tasks. We conducted a Fitts' law experiment with 41 subjects comparing head pointing and gaze pointing using a 300 ms dwell (<i>n</i> = 22) or click (<i>n</i> = 19) activation, with mouse input providing a baseline for both conditions. Gaze and head pointing were equally fast but slower than the mouse; dwell activation was faster than click activation. Throughput was highest for the mouse (2.75 bits/s), followed by head pointing (2.04 bits/s) and gaze pointing (1.85 bits/s). With dwell activation, however, throughput for gaze and head pointing were almost identical, as was the effective target width (≈ 55 pixels; about 2°) for all three input methods. Subjective feedback rated the physical workload less for gaze pointing than head pointing.
Original languageEnglish
Title of host publicationProceedings of the Workshop on Communication by Gaze Interaction
PublisherAssociation for Computing Machinery
Publication date2018
Pages1-5
Article numberArticle No. 7
ISBN (print)978-1-4503-5790-6
DOIs
StatePublished - 2018
EventWorkshop on Communication by Gaze Interaction - Warsaw, Poland

Conference

ConferenceWorkshop on Communication by Gaze Interaction
CountryPoland
CityWarsaw
Period15/06/201815/06/2018

  PDF

Slice-wise motion tracking during simultaneous EEG-fMRI

År: 2018

Slice-wise motion tracking during simultaneous EEG-fMRI

Laustsen, M., Andersen, M., Lehmann, P. M., Xue, R., Madsen, K. H. & Hanson, L. G. 2018

Publication: Research - peer-reviewConference abstract for conference – Annual report year: 2018

Slice-wise motion tracking during combined electroencephalography (EEG) and echo planar imaging (EPI) is developed. Using gradient-induced noise on the EEG for tracking, no interleaved navigator modules or additional hardware is needed. The motion parameters are determined after a calibration and training scan. The method is explored in a phantom and in vivo.
Original languageEnglish
Publication date2018
Number of pages2
StatePublished - 2018
EventJoint Annual Meeting ISMRM-ESMRMB 2018 - Paris, France

Conference

ConferenceJoint Annual Meeting ISMRM-ESMRMB 2018
LocationParis Expo Porte de Versailles
CountryFrance
CityParis
Period16/06/201821/06/2018
Internet address

  PDF

Academic performance and behavioral patterns

År: 2018

Academic performance and behavioral patterns

Kassarnig, V., Mones, E., Bjerre-Nielsen, A., Sapiezynski, P., Dreyer Lassen, D. & Lehmann, S. 2018 In : Epj Data Science. 7, 1, p. 1-16 16 p.

Publication: Research - peer-reviewJournal article – Annual report year: 2018

Identifying the factors that influence academic performance is an essential part of educational research. Previous studies have documented the importance of personality traits, class attendance, and social network structure. Because most of these analyses were based on a single behavioral aspect and/or small sample sizes, there is currently no quantification of the interplay of these factors. Here, we study the academic performance among a cohort of 538 undergraduate students forming a single, densely connected social network. Our work is based on data collected using smartphones, which the students used as their primary phones for two years. The availability of multi-channel data from a single population allows us to directly compare the explanatory power of individual and social characteristics. We find that the most informative indicators of performance are based on social ties and that network indicators result in better model performance than individual characteristics (including both personality and class attendance). We confirm earlier findings that class attendance is the most important predictor among individual characteristics. Finally, our results suggest the presence of strong homophily and/or peer effects among university students.
Original languageEnglish
JournalEpj Data Science
Volume7
Issue number1
Pages (from-to)1-16
Number of pages16
ISSN2193-1127
DOIs
StatePublished - 2018

  PDF

A Vibrotactile Alarm System for Pleasant Awakening

År: 2018

A Vibrotactile Alarm System for Pleasant Awakening

Korres, G., Jensen, C. B. F., Park, W., Bartsch, C. & Eid, M. 2018 In : Ieee Transactions on Haptics. 11 p.

Publication: Research - peer-reviewJournal article – Annual report year: 2018

There has been a vast development of personal informatics devices combining sleep monitoring with alarm systems, in order to find an optimal time to awaken a sleeping person in a pleasant way. Most of these systems implement auditory feedback, which is not always pleasant and may disturb other sleepers. We present an adaptive alarm system that detects sleeping cycles and triggers alarm signal during shallow sleep, to minimize sleep inertia. Since tactile sensation is associated with positive valence, vibrotactile stimulation is investigated as a silent alarm to enhance pleasant awakening. Three modulation techniques to render the tactile stimuli for pleasant awakening are considered, namely simultaneous, continuous, and successive stimulation. Two experimental studied are conducted. Experiment 1 studied exogenous attention towards tactile stimulation in a multimodal scenario (involving visual and haptic interactions) with fully awake individuals. Results from the attention task and the subjective valence rating suggest that the vibrotactile stimulation should be based on the continuous modulation, since this not only is very perceivable but also associated with positive attention. Experiment 2 evaluated the user experience with tactile stimulation patterns during sleep. Results confirmed the findings of experiment 1. Continuous modulation was rated highest for pleasant yet arousing sleep-awake transition.
Original languageEnglish
JournalIeee Transactions on Haptics
Number of pages11
ISSN2329-4051
DOIs
StateAccepted/In press - 2018

 

Perspectives on Machine Learning for Classification of Schizotypy Using fMRI Data

År: 2018

Perspectives on Machine Learning for Classification of Schizotypy Using fMRI Data

Madsen, K. H., Krohne, L. G., Cai, X-L., Wang, Y. & Chan, R. C. K. 2018 In : Schizophrenia Bulletin. 11 p.

Publication: Research - peer-reviewJournal article – Annual report year: 2018

Functional magnetic resonance imaging is capable of estimating functional activation and connectivity in the human brain, and lately there has been increased interest in the use of these functional modalities combined with machine learning for identification of psychiatric traits. While these methods bear great potential for early diagnosis and better understanding of disease processes, there are wide ranges of processing choices and pitfalls that may severely hamper interpretation and generalization performance unless carefully considered. In this perspective article, we aim to motivate the use of machine learning schizotypy research. To this end, we describe common data processing steps while commenting on best practices and procedures. First, we introduce the important role of schizotypy to motivate the importance of reliable classification, and summarize existing machine learning literature on schizotypy. Then, we describe procedures for extraction of features based on fMRI data, including statistical parametric mapping, parcellation, complex network analysis, and decomposition methods, as well as classification with a special focus on support vector classification and deep learning. We provide more detailed descriptions and software as supplementary material. Finally, we present current challenges in machine learning for classification of schizotypy and comment on future trends and perspectives.
Original languageEnglish
JournalSchizophrenia Bulletin
Number of pages11
ISSN0586-7614
DOIs
StatePublished - 2018

 

Machine Learning meets Mathematical Optimization to predict the optimal production of offshore wind parks

År: 2018

In this paper we propose a combination of Mathematical Optimization and Machine Learning to estimate the value of optimized solutions. In particular, we investigate if a machine, trained on a large number of optimized solutions, could accurately estimate the value of the optimized solution for new instances. In this paper we will focus on a specific application: the offshore wind farm layout optimization problem. Mixed Integer Programming models and other state-of-the-art optimization techniques, have been developed to solve this problem. Given the complexity of the problem and the big difference in production between optimized/non optimized solutions, it is not trivial to understand the potential value of a new site without running a complete optimization. This could be too time consuming if a lot of sites need to be evaluated, therefore we propose to use Machine Learning to quickly estimate the potential of new sites (i.e., to estimate the optimized production of a site without explicitly running the optimization). To do so, we trained and tested different Machine Learning models on a dataset of 3000+ optimized layouts found by the optimizer. Thanks to the close collaboration with a leading company in the energy sector, our model was trained on real-world data. Our results show that Machine Learning is able to efficiently estimate the value of optimized instances for the offshore wind farm layout problem
Original languageEnglish
JournalComputers & Operations Research
Number of pages27
ISSN0305-0548
DOIs
StatePublished - 2018

  PDF

Functional neuroimaging of recovery from motor conversion disorder: A case report: A case report

År: 2018

Functional neuroimaging of recovery from motor conversion disorder: A case report: A case report

Dogonowski, A. M., Andersen, K. W., Sellebjerg, F., Schreiber, K., Madsen, K. H. & Siebner, H. R. 2018 In : Neuroimage. p. 1-6

Publication: Research - peer-reviewJournal article – Annual report year: 2018

A patient with motor conversion disorder presented with a functional paresis of the left hand. After exclusion of structural brain damage, she was repeatedly examined with whole-brain functional magnetic resonance imaging, while she performed visually paced finger-tapping tasks. The dorsal premotor cortex showed a bilateral deactivation in the acute-subacute phase. Recovery from unilateral hand paresis was associated with a gradual increase in task-based activation of the dorsal premotor cortex bilaterally. The right medial prefrontal cortex displayed the opposite pattern, showing initial task-based activation that gradually diminished with recovery. The inverse dynamics of premotor and medial prefrontal activity over time were found during unimanual finger-tapping with the affected and non-affected hand as well as during bimanual finger-tapping. These observations suggest that reduced premotor and increased medial prefrontal activity reflect an effector-independent cortical dysfunction in conversion paresis which gradually disappears in parallel with clinical remission of paresis. The results link the medial prefrontal and dorsal premotor areas to the generation of intentional actions. We hypothesise that an excessive 'veto' signal generated in medial prefrontal cortex along with decreased premotor activity might constitute the functional substrate of conversion disorder. This notion warrants further examination in a larger group of affected patients.
Original languageEnglish
JournalNeuroimage
Pages (from-to)1-6
ISSN1053-8119
DOIs
StateAccepted/In press - 2018

  PDF

Frontal Brain Asymmetry and Willingness to Pay

År: 2018

Frontal Brain Asymmetry and Willingness to Pay

Ramsøy, T. Z., Skov, M., Christensen, M. K. & Stahlhut, C. 2018 In : Frontiers in Neuroscience. 12, 12 p., Article 138

Publication: Research - peer-reviewJournal article – Annual report year: 2018

Consumers frequently make decisions about how much they are willing to pay (WTP) for specific products and services, but little is known about the neural mechanisms underlying such calculations. In this study, we were interested in testing whether specific brain activation-the asymmetry in engagement of the prefrontal cortex-would be related to consumer choice. Subjects saw products and subsequently decided how much they were willing to pay for each product, while undergoing neuroimaging using electroencephalography. Our results demonstrate that prefrontal asymmetry in the gamma frequency band, and a trend in the beta frequency band that was recorded during product viewing was significantly related to subsequent WTP responses. Frontal asymmetry in the alpha band was not related to WTP decisions. Besides suggesting separate neuropsychological mechanisms of consumer choice, we find that one specific measure-the prefrontal gamma asymmetry-was most strongly related to WTP responses, and was most coupled to the actual decision phase. These findings are discussed in light of the psychology of WTP calculations, and in relation to the recent emergence of consumer neuroscience and neuromarketing.
Original languageEnglish
Article numberArticle 138
JournalFrontiers in Neuroscience
Volume12
Number of pages12
ISSN1662-4548
DOIs
StatePublished - 2018

 

Deep Latent Variable Models for Sequential Data

År: 2018

Deep Latent Variable Models for Sequential Data

Fraccaro, M. 2018 DTU Compute. 146 p. (DTU Compute PHD-2018, Vol. 475).

Publication: ResearchPh.D. thesis – Annual report year: 2018

Over the last few decades an ever-increasing amount of data is being collected in a wide range of applications. This has boosted the development of mathematical models that are able to analyze it and discover its underlying structure, and use the extracted information to solve a multitude of different tasks, such as for predictive modelling or pattern recognition. The available data is however often complex and high-dimensional, making traditional data analysis methods ineffective in many applications. In the recent years there has then been a big focus on the development of more powerful models, that need to be general enough to be able to handle many diverse applications and kinds of data. Some of the most interesting advancements in this research direction have been recently obtained combining ideas from probabilistic modelling and deeplearning. Variationalauto-encoders(VAEs), that belong to the broader family of deep latent variable models, are powerful and scalable models that can be used for unsupervised learning of complex high-dimensional data distributions. They achieve this by parameterizing expressive probability distributions over the latent variables of the model using deep neural networks. VAEs can be used in applications with static data, for example as a generative model of images, but they are not suitable to model temporal data such as the sequences of images that form a video. However, a major part of the data that is being collected has a sequential nature, and finding powerful architectures that are able to model it is therefore fundamental. In the first part of the thesis we will introduce a broad class of deep latent variable models for sequential data, that can be used for unsupervised learning of complex and high-dimensional sequential data distributions. We obtain these models by extending VAEs to the temporal setting, and further combining ideas from deep learning (e.g. deep and recurrent neural networks) and probabilistic modelling (e.g. state-space models) to define generative models for the data that use deep neural networks to parameterize very flexible probability distributions. This results in a family of powerful architectures that can model a wide range of complex temporal data, and can be trained in a scalable way using large unlabelled datasets. In the second part of the thesis we will then present in detail three architectures belonging to this family of models. First, we will introduce stochastic recurrent neural networks (Fraccaro et al., 2016c), that combine the expressiveness of recurrent neural networks and the ability of state-space models to model the uncertainty in the learned latent representation. We will then present Kalman variational auto-encoders (Fraccaro et al., 2017), that can learn from data disentangled and more interpretable visual and dynamic representations. Finally, we will show that to deal with temporal applications that require a high memory capacity we can combine deep latent variable models with external memory architectures, as in the generative temporal model with spatial memory of Fraccaro et al. (2018).
Original languageEnglish
PublisherDTU Compute
Number of pages146
StateSubmitted - 2018
SeriesDTU Compute PHD-2018
Volume475
ISSN0909-3192

 

Spatio-temporal methods for EEG analysis in cognitive neuroscience

År: 2018

Spatio-temporal methods for EEG analysis in cognitive neuroscience

Poulsen, A. T. 2018 DTU Compute. 152 p. (DTU Compute PHD-2018, Vol. 473).

Publication: ResearchPh.D. thesis – Annual report year: 2018

Electroencephalography (EEG) records electrical activity from the brain by measuring the resulting potential differences across the scalp. It has a long tradition in both a clinical and neuroscientific setting, and recently it has also started being used for consumer-oriented applications. While EEG can be a useful tool, it can be difficult to decipher information from its raw signals. In this thesis I will present three projects with the common goal of analysing EEG in ways that both extract meaningful information and visualise it in intuitive ways. The first project describes how we took neuroscience out of the laboratory and into the classroom. We reproduced an attention-tracking paradigm in a classroom and simultaneously recorded the neural activity of up to nine people. We had a focus on using equipment that was wireless and portable as well being relatively low-cost and computational methods in a setup that is feasible to extend into everyday scenarios. The second project revolved around creating a toolbox for the research field of microstate analysis, with a focus on open access and transparency of the applied methods. The toolbox is followed by a methodological guide that reviews the most commonly applied algorithms in microstate analysis. In the final project I investigated the feasibility of using the complexity of EEG as a neural marker of conscious processing. This project spans two studies investigating the capability of EEG complexity in two different scenarios; while people are sleeping, and while navigating a helicopter simulator.
Original languageEnglish
PublisherDTU Compute
Number of pages152
StateSubmitted - 2018
SeriesDTU Compute PHD-2018
Volume473
ISSN0909-3192

  PDF

Deep Generative Models for Molecular Science

År: 2018

Deep Generative Models for Molecular Science

Jørgensen, P. B., Schmidt, M. N. & Winther, O. 2018 In : Molecular Informatics. 37, 1-2, 9 p., 1700133

Publication: Research - peer-reviewJournal article – Annual report year: 2018

Generative deep machine learning models now rival traditional quantum-mechanical computations in predicting properties of new structures, and they come with a significantly lower computational cost, opening new avenues in computational molecular science. In the last few years, a variety of deep generative models have been proposed for modeling molecules, which differ in both their model structure and choice of input features. We review these recent advances within deep generative models for predicting molecular properties, with particular focus on models based on the probabilistic autoencoder (or variational autoencoder, VAE) approach in which the molecular structure is embedded in a latent vector space from which its properties can be predicted and its structure can be restored.
Original languageEnglish
Article number1700133
JournalMolecular Informatics
Volume37
Issue number1-2
Number of pages9
ISSN1868-1743
DOIs
StatePublished - 2018

 

Automatic skull segmentation from MR images for realistic volume conductor models of the head: Assessment of the state-of-the-art

År: 2018

Automatic skull segmentation from MR images for realistic volume conductor models of the head: Assessment of the state-of-the-art

Nielsen, J. D., Madsen, K. H., Puonti, O., Siebner, H. R., Bauer, C., Madsen, C. G., Saturnino, G. B. & Thielscher, A. 2018 In : Neuroimage. 12 p.

Publication: Research - peer-reviewJournal article – Annual report year: 2018

Anatomically realistic volume conductor models of the human head are important for accurate forward modeling of the electric field during transcranial brain stimulation (TBS), electro- (EEG) and magnetoencephalography (MEG). In particular, the skull compartment exerts a strong influence on the field distribution due to its low conductivity, suggesting the need to represent its geometry accurately. However, automatic skull reconstruction from structural magnetic resonance (MR) images is difficult, as compact bone has a very low signal in magnetic resonance imaging (MRI). Here, we evaluate three methods for skull segmentation, namely FSL BET2, the unified segmentation routine of SPM12 with extended spatial tissue priors, and the skullfinder tool of BrainSuite. To our knowledge, this study is the first to rigorously assess the accuracy of these state-of-the-art tools by comparison with CT-based skull segmentations on a group of ten subjects. We demonstrate several key factors that improve the segmentation quality, including the use of multi-contrast MRI data, the optimization of the MR sequences and the adaptation of the parameters of the segmentation methods. We conclude that FSL and SPM12 achieve better skull segmentations than BrainSuite. The former methods obtain reasonable results for the upper part of the skull when a combination of T1- and T2-weighted images is used as input. The SPM12-based results can be improved slightly further by means of simple morphological operations to fix local defects. In contrast to FSL BET2, the SPM12-based segmentation with extended spatial tissue priors and the BrainSuite-based segmentation provide coarse reconstructions of the vertebrae, enabling the construction of volume conductor models that include the neck. We exemplarily demonstrate that the extended models enable a more accurate estimation of the electric field distribution during transcranial direct current stimulation (tDCS) for montages that involve extraencephalic electrodes. The methods provided by FSL and SPM12 are integrated into pipelines for the automatic generation of realistic head models based on tetrahedral meshes, which are distributed as part of the open-source software package SimNIBS for field calculations for transcranial brain stimulation.
Original languageEnglish
JournalNeuroimage
Number of pages12
ISSN1053-8119
DOIs
StatePublished - 2018

 

Deep Generative Models for Semi-Supervised Machine Learning

År: 2018

Deep Generative Models for Semi-Supervised Machine Learning

Maaløe, L. 2018 DTU Compute. 155 p. (DTU Compute PHD-2018, Vol. 472).

Publication: ResearchPh.D. thesis – Annual report year: 2018

The reintroduction of deep neural networks has a large impact on the modeling capabilities of modern machine learning. This reignites the general public’s dream of achieving artificial intelligence, and spawns rapid progress in large scale industrial machine learning development, such as autonomous driving. However, the leaps in development are still confined to a rather limited learning domain, in which labeled data is required. Labeled data is hard and costly to acquire, due to the amount needed to efficiently learn a modern machine learning model, and that many data sources are not directly interpretable. Consequently, research in different learning paradigms that utilize vast amounts of unlabeled data is getting more and more attention. Albeit possessing intriguing theoretical properties, machine learning models that learn from unlabeled data are still an unsolved research topic. The thesis comprises methods that utilize the power of deep neural networks to learn from both labeled and unlabeled data. A background for the theoretical foundation of the proposed methods are described and empirical results showing their capabilities within generation and classification tasks are presented. Finally, a real-life application within condition monitoring for sustainable energy is demonstrated, proving that the proposed methods have the expected impact and are applicable.
Original languageEnglish
PublisherDTU Compute
Number of pages155
StateSubmitted - 2018
SeriesDTU Compute PHD-2018
Volume472
ISSN0909-3192

 

Functional Connectivity using a Wishart Mixture Model

År: 2018

Functional Connectivity using a Wishart Mixture Model

Nielsen, S. F. V., Madsen, K. H., Schmidt, M. N. & Mørup, M. 2018 Proceedings of International Workshop on Pattern Recognition in Neuroimaging .

Publication: Research - peer-reviewArticle in proceedings – Annual report year: 2018

Original languageEnglish
Title of host publicationProceedings of International Workshop on Pattern Recognition in Neuroimaging
Publication date2018
StatePublished - 2018
Event2017 International Workshop on Pattern Recognition in Neuroimaging - Toronto, Canada

Conference

Conference2017 International Workshop on Pattern Recognition in Neuroimaging
LocationUniversity of Toronto
CountryCanada
CityToronto
Period21/06/201723/06/2017

 

Altered auditory processing and effective connectivity in 22q11.2 deletion syndrome

År: 2018

Altered auditory processing and effective connectivity in 22q11.2 deletion syndrome

Larsen, K. M., Mørup, M., Birknow, M. R., Fischer, E., Hulme, O., Vangkilde, A., Schmock, H., Baaré, W. F. C., Didriksen, M., Olsen, L., Werge, T., Siebner, H. R. & Garrido, M. I. 2018 In : Schizophrenia Research. 9 p.

Publication: Research - peer-reviewJournal article – Annual report year: 2018

22q11.2 deletion syndrome (22q11.2DS) is one of the most common copy number variants and confers a markedly increased risk for schizophrenia. As such, 22q11.2DS is a homogeneous genetic liability model which enables studies to delineate functional abnormalities that may precede disease onset. Mismatch negativity (MMN), a brain marker of change detection, is reduced in people with schizophrenia compared to healthy controls. Using dynamic causal modelling (DCM), previous studies showed that top-down effective connectivity linking the frontal and temporal cortex is reduced in schizophrenia relative to healthy controls in MMN tasks. In the search for early risk-markers for schizophrenia we investigated the neural basis of change detection in a group with 22q11.2DS. We recorded high-density EEG from 19 young non-psychotic 22q11.2 deletion carriers, as well as from 27 healthy non-carriers with comparable age distribution and sex ratio, while they listened to a sequence of sounds arranged in a roving oddball paradigm. Despite finding no significant reduction in the MMN responses, whole-scalp spatiotemporal analysis of responses to the tones revealed a greater fronto-temporal N1 component in the 22q11.2 deletion carriers. DCM showed reduced intrinsic connection within right primary auditory cortex as well as in the top-down, connection from the right inferior frontal gyrus to right superior temporal gyrus for 22q11.2 deletion carriers although not surviving correction for multiple comparison. We discuss these findings in terms of reduced adaptation and a general increased sensitivity to tones in 22q11.2DS.

Original languageEnglish
JournalSchizophrenia Research
Number of pages9
ISSN0920-9964
DOIs
StatePublished - 2018

  PDF

Understanding predictability and exploration in human mobility

År: 2018

Understanding predictability and exploration in human mobility

Cuttone, A., Jørgensen, S. L. & González, M. C. 2018 In : Epj Data Science. 7, 2, p. 1-17 17 p.

Publication: Research - peer-reviewJournal article – Annual report year: 2018

Predictive models for human mobility have important applications in many fields including traffic control, ubiquitous computing, and contextual advertisement. The predictive performance of models in literature varies quite broadly, from over 90% to under 40%. In this work we study which underlying factors - in terms of modeling approaches and spatio-temporal characteristics of the data sources - have resulted in this remarkably broad span of performance reported in the literature. Specifically we investigate which factors influence the accuracy of next-place prediction, using a high-precision location dataset of more than 400 users observed for periods between 3 months and one year. We show that it is much easier to achieve high accuracy when predicting the time-bin location than when predicting the next place. Moreover, we demonstrate how the temporal and spatial resolution of the data have strong influence on the accuracy of prediction. Finally we reveal that the exploration of new locations is an important factor in human mobility, and we measure that on average 20-25% of transitions are to new places, and approx. 70% of locations are visited only once. We discuss how these mechanisms are important factors limiting our ability to predict human mobility.
Original languageEnglish
JournalEpj Data Science
Volume7
Issue number2
Pages (from-to)1-17
Number of pages17
ISSN2193-1127
DOIs
StatePublished - 2018

Bibliographical note

© The Author(s) 2018. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made

 

Benefits of spatio-temporal modelling for short term wind power forecasting at both individual and aggregated levels

År: 2018

Benefits of spatio-temporal modelling for short term wind power forecasting at both individual and aggregated levels

Lenzi, A., Pinson, P. & Steinsland, I. 2018 In : Environmetrics. 35 p.

Publication: Research - peer-reviewJournal article – Annual report year: 2018

The share of wind energy in total installed power capacity has grown rapidly in recent years. Producing accurate and reliable forecasts of wind power production, together with a quantification of the uncertainty, is essential to optimally integrate wind energy into power systems. We build spatio-temporal models for wind power generation and obtain full probabilistic forecasts from 15 minutes to 5 hours ahead. Detailed analysis of the forecast performances on the individual wind farms and aggregated wind power are provided. The predictions from our models are evaluated on a data set from wind farms in western Denmark using a sliding window approach, for which estimation is performed using only the last available measurements. The case study shows that it is important to have a spatio-temporal model instead of a temporal one to achieve calibrated aggregated forecasts. Furthermore, spatio-temporal models have the advantage of being able to produce spatially out-of-sample forecasts. We use a Bayesian hierarchical framework to obtain fast and accurate forecasts of wind power generation at wind farms where recent data are available, but also at a larger portfolio including wind farms without recent observations of power production.The results and the methodologies are relevant for wind power forecasts across the globe as well as forspatial-temporal modelling in general.
Original languageEnglish
JournalEnvironmetrics
Number of pages35
ISSN1180-4009
StateSubmitted - 2018

 

Optimizing targeted vaccination across cyber-physical networks: an empirically based mathematical simulation study

År: 2018

Optimizing targeted vaccination across cyber-physical networks: an empirically based mathematical simulation study

Mones, E., Stopczynski, A., Pentland, A. . S., Hupert, N. & Lehmann, S. 2018 In : Journal of the Royal Society. Interface. 15, 138, 20170783

Publication: Research - peer-reviewJournal article – Annual report year: 2018

Targeted vaccination, whether to minimize the forward transmission of infectious diseases or their clinical impact, is one of the 'holy grails' of modern infectious disease outbreak response, yet it is difficult to achieve in practice due to the challenge of identifying optimal targets in real time. If interruption of disease transmission is the goal, targeting requires knowledge of underlying person-to-person contact networks. Digital communication networks may reflect not only virtual but also physical interactions that could result in disease transmission, but the precise overlap between these cyber and physical networks has never been empirically explored in real-life settings. Here, we study the digital communication activity of more than 500 individuals along with their person-to-person contacts at a 5-min temporal resolution. We then simulate different disease transmission scenarios on the person-to-person physical contact network to determine whether cyber communication networks can be harnessed to advance the goal of targeted vaccination for a disease spreading on the network of physical proximity. We show that individuals selected on the basis of their closeness centrality within cyber networks (what we call 'cyber-directed vaccination') can enhance vaccination campaigns against diseases with short-range (but not full-range) modes of transmission.
Original languageEnglish
Article number20170783
JournalJournal of the Royal Society. Interface
Volume15
Issue number138
ISSN1742-5689
DOIs
StatePublished - 2018

 

Predictive assessment of models for dynamic functional connectivity

År: 2018

Predictive assessment of models for dynamic functional connectivity

Nielsen, S. F. V., Schmidt, M. N., Madsen, K. H. & Mørup, M. 2018 In : Neuroimage. 171, 35 p.

Publication: Research - peer-reviewJournal article – Annual report year: 2017

In neuroimaging, it has become evident that models of dynamic functional connectivity (dFC), which characterize how intrinsic brain organization changes over time, can provide a more detailed representation of brain function than traditional static analyses. Many dFC models in the literature represent functional brain networks as a meta-stable process with a discrete number of states; however, there is a lack of consensus on how to perform model selection and learn the number of states, as well as a lack of understanding of how different modeling assumptions influence the estimated state dynamics. To address these issues, we consider a predictive likelihood approach to model assessment, where models are evaluated based on their predictive performance on held-out test data. Examining several prominent models of dFC (in their probabilistic formulations) we demonstrate our framework on synthetic data, and apply it on two real-world examples: a face recognition EEG experiment and resting-state fMRI. Our results evidence that both EEG and fMRI are better characterized using dynamic modeling approaches than by their static counterparts, but we also demonstrate that one must be cautious when interpreting dFC because parameter settings and modeling assumptions, such as window lengths and emission models, can have a large impact on the estimated states and consequently on the interpretation of the brain dynamics.
Original languageEnglish
JournalNeuroimage
Volume171
Number of pages35
ISSN1053-8119
DOIs
StatePublished - 2018

 

Deep learning for automated drivetrain fault detection

År: 2018

Deep learning for automated drivetrain fault detection

Bach-Andersen, M., Rømer-Odgaard, B. & Winther, O. 2018 In : Wind Energy. 21, 1, p. 29-41 13 p.

Publication: Research - peer-reviewJournal article – Annual report year: 2018

A novel data-driven deep-learning system for large-scale wind turbine drivetrain monitoring applications is presented. It uses convolutional neural network processing on complex vibration signal inputs. The system is demonstrated to learn successfully from the actions of human diagnostic experts and provide early and robust fault detection on both rotor bearing, planetary and helical stage gear box bearings from analysis of multisensor vibration patterns using only a high-level feature selection. On the basis of data from 251 actual wind turbine bearing failures, we are able to accurately quantify the fleet-wide diagnostic model performance. The analysis also explores the time dependence of the diagnostic performance, providing a detailed view of the timeliness and accuracy of the diagnostic outputs across the different architectures. Deep architectures are shown to outperform the human analyst as well as shallow-learning architectures, and the results demonstrate that when applied in a large-scale monitoring system, machine intelligence is now able to handle some of the most challenging diagnostic tasks related to wind turbines.
Original languageEnglish
JournalWind Energy
Volume21
Issue number1
Pages (from-to)29-41
Number of pages13
ISSN1095-4244
DOIs
StatePublished - 2018

 

The circle equation over finite fields

År: 2018

The circle equation over finite fields

Aabrandt, A. & Hansen, V. L. 2018 In : Quaestiones Mathematicae. 41, 5, p. 665-674 10 p.

Publication: Research - peer-reviewJournal article – Annual report year: 2018

Interesting patterns in the geometry of a plane algebraic curve C can be observed when the defining polynomial equation is solved over the family of finite fields. In this paper, we examine the case of C the classical unit circle defined by the circle equation x2 + y2 = 1. As a main result, we establish a concise formula for the number of solutions to the circle equation over an arbitrary finite field. We also provide criteria for the existence of diagonal solutions to the circle equation. Finally, we give a precise description of how the number of solutions to the circle equation over a prime field grows as a function of the prime.

Original languageEnglish
JournalQuaestiones Mathematicae
Volume41
Issue number5
Pages (from-to)665-674
Number of pages10
ISSN1607-3606
DOIs
StatePublished - 2018

  PDF

Task-Modulated Cortical Representations of Natural Sound Source Categories

År: 2018

Task-Modulated Cortical Representations of Natural Sound Source Categories

Hjortkjær, J., Kassuba, T., Madsen, K. H., Skov, M. & Siebner, H. R. 2018 In : Cerebral Cortex. 128, 1, p. 295-306 12 p.

Publication: Research - peer-reviewJournal article – Annual report year: 2018

In everyday sound environments, we recognize sound sources and events by attending to relevant aspects of an acoustic input. Evidence about the cortical mechanisms involved in extracting relevant category information from natural sounds is, however, limited to speech. Here, we used functional MRI to measure cortical response patterns while human listeners categorized real-world sounds created by objects of different solid materials (glass, metal, wood) manipulated by different sound-producing actions (striking, rattling, dropping). In different sessions, subjects had to identify either material or action categories in the same sound stimuli. The sound-producing action and the material of the sound source could be decoded from multivoxel activity patterns in auditory cortex, including Heschl’s gyrus and planum temporale. Importantly, decoding success depended on task relevance and category discriminability. Action categories were more accurately decoded in auditory cortex when subjects identified action information. Conversely, the material of the same sound sources was decoded with higher accuracy in the inferior frontal cortex during material identification. Representational similarity analyses indicated that both early and higher-order auditory cortex selectively enhanced spectrotemporal features relevant to the target category. Together, the results indicate a cortical selection mechanism that favors task-relevant information in the processing of nonvocal sound categories.
Original languageEnglish
JournalCerebral Cortex
Volume128
Issue number1
Pages (from-to)295-306
Number of pages12
ISSN1047-3211
DOIs
StatePublished - 2018

  PDF

Design process robustness: A bi-partite network analysis reveals the central importance of people

År: 2018

Design process robustness: A bi-partite network analysis reveals the central importance of people

Piccolo, S., Jørgensen, S. L. & Maier, A. 2018 In : Design Science Journal. 4, e1

Publication: Research - peer-reviewJournal article – Annual report year: 2018

Design processes require the joint effort of many people to collaborate and work on multiple activities. Effective techniques to analyse and model design processes are important for understanding organisational dynamics, for improving collaboration, and for planning robust design processes, reducing the risk of rework and delays. Although there has been much progress in modelling and understanding design processes, little is known about the interplay between people and the activities they perform and its influence on design process robustness. To analyse this interplay, we model a large-scale design process of a biomass power plant with people and activities as a bipartite network. Observing that some people act as bridges between activities organised to form nearly independent modules, in order to evaluate process fragility, we simulate random failures and targeted attacks to people and activities. We find that our process is more vulnerable to attacks to people rather than activities. These findings show how the allocation of people to activities can obscure an inherent fragility, making the process highly sensitive and dependent on specific people. More generally, we show that the behaviour of robustness is determined by the degree distributions, the heterogeneity of which can be leveraged to improve robustness and resilience to cascading failures. Overall, we show that it is important to carefully plan the assignment of people to activities.
Original languageEnglish
Article numbere1
JournalDesign Science Journal
Volume4
DOIs
StatePublished - 2018

Bibliographical note

Distributed as Open Access under a CC-BY 4.0 license (http://creativecommons.org/licenses/by/4.0/)

  PDF

FindZebra - using machine learning to aid diagnosis of rare diseases

År: 2018

FindZebra - using machine learning to aid diagnosis of rare diseases

Svenstrup, D. T. 2018 DTU Compute. 118 p. (DTU Compute PHD-2017, Vol. 463).

Publication: ResearchPh.D. thesis – Annual report year: 2018

FindZebra is a search engine for rare diseases intended to act as a diagnosis decision support system (DDSS) capable of assisting the user both during and after a search. Rare diseases are diseases that affect only a small part of the population (less than one in two thousand). Currently around seven thousand rare diseases are known and it is estimated that 6−8% of the population will be affected by a rare disease during their lifetime. Due to their rarity and large number, diagnosis of rare diseases is difficult and often associated with year long delays and diagnostic errors. These difficulties with diagnosis have a profound human and societal cost. This means that even a small increase in success rate when using a tool such as FindZebra could potentially have a great impact on society. In this dissertation we explore four lines of research for improving FindZebra using machine learning methods. The first line of research is on how to improve the retrieval performance of FindZebra. By using a combination of improved models, medical databases and corpus expansion we show that it is possible to obtain a substantial improvement in retrieval performance compared to current state-of-the-art document retrieval systems. Improving retrieval performance is important, but is not the only way of improving the success rate of a DDSS such as FindZebra. Following an unsuccessful search, the search engine should assist the user by indicating what information is likely to be missing. This idea is called Information Completion (IC) and will be explored in the second line of research. In order to represent words (and other discrete tokens) in a neural network it is necessary to transform each word to a vector form. This is typically accomplished by using a word embedding, which is an essential component in any word based neural network. The third line of research is on how to improve this basic component. Users of FindZebra who do not have English as their primary language often have difficulty expressing complex medical queries in English. Optimally, a user should be able to write a query in his or her native language and the search engine should then give a suggestion for a differential diagnosis based on all the information contained in a multilingual corpus, not only in the native corpus. Methods for performing multilingual search will be the fourth line of research explored in this dissertation. 

Original languageEnglish
PublisherDTU Compute
Number of pages118
StatePublished - 2018
SeriesDTU Compute PHD-2017
Volume463
ISSN0909-3192

  PDF

A Diagnostic and Predictive Framework for Wind Turbine Drive Train Monitoring

År: 2018

A Diagnostic and Predictive Framework for Wind Turbine Drive Train Monitoring

Bach-Andersen, M., Winther, O. & Rømer-Odgaard, B. 2018 Technical University of Denmark (DTU). 111 p. (DTU Compute PHD-2017, Vol. 449).

Publication: ResearchPh.D. thesis – Annual report year: 2018

Vast amount of data are collected minute by minute from wind turbines around the world. This thesis represents a focused research effort into discovering new ways of processing these data streams in order to gain insights which can be used to lower the maintenance costs of wind turbines and increase the turbine availability.
First, it is demonstrated how simple sensor data streams can be leveraged based on a combination of non-linear predictive models and unsupervised fault detection to provide warnings of a critical bearing failure more than a month earlier compared to existing alarm systems. Second, early fault identification based on analysis of complex vibration patterns which is a domain previously reserved for human experts, is shown to be solved with high accuracy using deep learning architecture strained in a fully supervised sense from the data collected in a large scale wind turbine monitoring platform. The research shows a way towards a fully automatized data-driven wind turbine diagnostic processing system that is highly scalable and requires little or no feature engineering and system modeling.
Original languageEnglish
PublisherTechnical University of Denmark (DTU)
Number of pages111
StatePublished - 2018
SeriesDTU Compute PHD-2017
Volume449
ISSN0909-3192

 

The Choice of Prior in Bayesian Modeling of the Information Sampling Task

År: 2018

The Choice of Prior in Bayesian Modeling of the Information Sampling Task

Axelsen, M. C., Jepsen, J. R. M. & Bak, N. 2018 In : Biological Psychiatry. 83, 12, p. E59-E60 2 p.

Publication: ResearchLetter – Annual report year: 2018

Original languageEnglish
JournalBiological Psychiatry
Volume83
Issue number12
Pages (from-to)E59-E60
Number of pages2
ISSN0006-3223
DOIs
StatePublished - 2018

2017
 

How to target inter-regional phase synchronization with dual-site Transcranial Alternating Current Stimulation

År: 2017

How to target inter-regional phase synchronization with dual-site Transcranial Alternating Current Stimulation

Saturnino, G. B., Madsen, K. H., Siebner, H. R. & Thielscher, A. 1 Dec 2017 In : NeuroImage. 163, p. 68-80

Publication: Research - peer-reviewJournal article – Annual report year: 2017

Large-scale synchronization of neural oscillations is a key mechanism for functional information exchange among brain areas. Dual-site Transcranial Alternating Current Stimulation (ds-TACS) has been recently introduced as non-invasive technique to manipulate the temporal phase relationship of local oscillations in two connected cortical areas. While the frequency of ds-TACS is matched, the phase of stimulation is either identical (in-phase stimulation) or opposite (anti-phase stimulation) in the two cortical target areas. In-phase stimulation is thought to synchronize the endogenous oscillations and hereby to improve behavioral performance. Conversely, anti-phase stimulation is thought to desynchronize neural oscillations in the two areas, which is expected to decrease performance. Critically, in- and anti-phase ds-TACS should only differ with respect to temporal phase, while all other stimulation parameters such as focality and stimulation intensity should be matched to enable an unambiguous interpretation of the behavioral effects. Using electric field simulations based on a realistic head geometry, we tested how well this goal has been met in studies, which have employed ds-TACS up to now. Separating the induced electrical fields in their spatial and temporal components, we investigated how the chosen electrode montages determined the spatial field distribution and the generation of phase variations in the injected electric fields. Considering the basic physical mechanisms, we derived recommendations for an optimized stimulation montage. The latter allows for a principled design of in- and anti-phase ds-TACS conditions with matched spatial distributions of the electric field. This knowledge will help cognitive neuroscientists to design optimal ds-TACS configurations, which are suited to probe unambiguously the causal contribution of phase coupling to specific cognitive processes in the human brain.

Original languageEnglish
JournalNeuroImage
Volume163
Pages (from-to)68-80
ISSN1053-8119
DOIs
StatePublished - 1 Dec 2017

 

Flexible non-linear predictive models for large-scale wind turbine diagnostics

År: 2017

Flexible non-linear predictive models for large-scale wind turbine diagnostics

Bach-Andersen, M., Rømer-Odgaard, B. & Winther, O. 1 May 2017 In : Wind Energy. 20, 5, p. 753-764 12 p.

Publication: Research - peer-reviewJournal article – Annual report year: 2017

We demonstrate how flexible non-linear models can provide accurate and robust predictions on turbine component temperature sensor data using data-driven principles and only a minimum of system modeling. The merits of different model architectures are evaluated using data from a large set of turbines operating under diverse conditions. We then go on to test the predictive models in a diagnostic setting, where the output of the models are used to detect mechanical faults in rotor bearings. Using retrospective data from 22 actual rotor bearing failures, the fault detection performance of the models are quantified using a structured framework that provides the metrics required for evaluating the performance in a fleet wide monitoring setup. It is demonstrated that faults are identified with high accuracy up to 45 days before a warning from the hard-threshold warning system.

Original languageEnglish
JournalWind Energy
Volume20
Issue number5
Pages (from-to)753-764
Number of pages12
ISSN1095-4244
DOIs
StatePublished - 1 May 2017

 

Maximum Likelihood Estimation of Riemannian Metrics from Euclidean Data

År: 2017

Maximum Likelihood Estimation of Riemannian Metrics from Euclidean Data

Arvanitidis, G., Hansen, L. K. & Hauberg, S. 2017 In : Lecture Notes in Computer Science. 10589, p. 38-46

Publication: Research - peer-reviewConference article – Annual report year: 2018

Euclidean data often exhibit a nonlinear behavior, which may be modeled by assuming the data is distributed near a nonlinear submanifold in the data space. One approach to find such a manifold is to estimate a Riemannian metric that locally models the given data. Data distributions with respect to this metric will then tend to follow the nonlinear structure of the data. In practice, the learned metric rely on parameters that are hand-tuned for a given task. We propose to estimate such parameters by maximizing the data likelihood under the assumed distribution. This is complicated by two issues: (1) a change of parameters imply a change of measure such that different likelihoods are incomparable; (2) some choice of parameters renders the numerical calculation of distances and geodesics unstable such that likelihoods cannot be evaluated. As a practical solution, we propose to (1) re-normalize likelihoods with respect to the usual Lebesgue measure of the data space, and (2) to bound the likelihood when its exact value is unattainable. We provide practical algorithms for these ideas and illustrate their use on synthetic data, images of digits and faces, as well as signals extracted from EEG scalp measurements.
Original languageEnglish
Book seriesLecture Notes in Computer Science
Volume10589
Pages (from-to)38-46
ISSN0302-9743
DOIs
StatePublished - 2017
Event3rd International Conference on Geometric Science of Information - Paris, France

Conference

Conference3rd International Conference on Geometric Science of Information
LocationGSI2017
CountryFrance
CityParis
Period07/11/201709/11/2017

  PDF

Dynamical Functional Theory for Compressed Sensing

År: 2017

Dynamical Functional Theory for Compressed Sensing

Cakmak, B., Opper, M., Winther, O. & Fleury, B. H. 2017 2017 IEEE International Symposium on Information Theory (ISIT). IEEE, p. 2143-2147 5 p.

Publication: ResearchArticle in proceedings – Annual report year: 2018

We introduce a theoretical approach for designing generalizations of the approximate message passing (AMP) algorithm for compressed sensing which are valid for large observation matrices that are drawn from an invariant random matrix ensemble. By design, the fixed points of the algorithm obey the Thouless-Anderson-Palmer (TAP) equations corresponding to the ensemble. Using a dynamical functional approach we are able to derive an effective stochastic process for the marginal statistics of a single component of the dynamics. This allows us to design memory terms in the algorithm in such a way that the resulting fields become Gaussian random variables allowing for an explicit analysis. The asymptotic statistics of these fields are consistent with the replica ansatz of the compressed sensing problem.
Original languageEnglish
Title of host publication2017 IEEE International Symposium on Information Theory (ISIT)
Number of pages5
PublisherIEEE
Publication date2017
Pages2143-2147
DOIs
StatePublished - 2017
Event2017 IEEE International Symposium on Information Theory - Aachen, Germany

Conference

Conference2017 IEEE International Symposium on Information Theory
CountryGermany
CityAachen
Period25/06/201730/06/2017

 

Deep convolutional neural networks for interpretable analysis of EEG sleep stage scoring

År: 2017

Deep convolutional neural networks for interpretable analysis of EEG sleep stage scoring

Vilamala, A., Madsen, K. H. & Hansen, L. K. 2017 Proceedings of the 2017 IEEE 27th International Workshop on Machine Learning for Signal Processing (MLSP). IEEE, 6 p.

Publication: ResearchArticle in proceedings – Annual report year: 2018

Sleep studies are important for diagnosing sleep disorders such as insomnia, narcolepsy or sleep apnea. They rely on manual scoring of sleep stages from raw polisomnography signals, which is a tedious visual task requiring the workload of highly trained professionals. Consequently, research efforts to purse for an automatic stage scoring based on machine learning techniques have been carried out over the last years. In this work, we resort to multitaper spectral analysis to create visually interpretable images of sleep patterns from EEG signals as inputs to a deep convolutional network trained to solve visual recognition tasks. As a working example of transfer learning, a system able to accurately classify sleep stages in new unseen patients is presented. Evaluations in a widely-used publicly available dataset favourably compare to state-of-the-art results, while providing a framework for visual interpretation of outcomes.
Original languageEnglish
Title of host publicationProceedings of the 2017 IEEE 27th International Workshop on Machine Learning for Signal Processing (MLSP)
Number of pages6
PublisherIEEE
Publication date2017
DOIs
StatePublished - 2017
Event27th International Workshop on Machine Learning for Signal Processing (MLSP) - Tokyo, Japan

Workshop

Workshop27th International Workshop on Machine Learning for Signal Processing (MLSP)
CountryJapan
CityTokyo
Period25/09/201728/09/2017

 

Adaptive Smoothing in fMRI Data Processing Neural Networks

År: 2017

Adaptive Smoothing in fMRI Data Processing Neural Networks

Vilamala, A., Madsen, K. H. & Hansen, L. K. 2017 Proceedings of the 2017 International Workshop on Pattern Recognition in Neuroimaging. IEEE, 4 p.

Publication: ResearchArticle in proceedings – Annual report year: 2018

Functional Magnetic Resonance Imaging (fMRI) relies on multi-step data processing pipelines to accurately determine brain activity; among them, the crucial step of spatial smoothing. These pipelines are commonly suboptimal, given the local optimisation strategy they use, treating each step in isolation. With the advent of new tools for deep learning, recent work has proposed to turn these pipelines into end-to-end learning networks. This change of paradigm offers new avenues to improvement as it allows for a global optimisation. The current work aims at benefitting from this paradigm shift by defining a smoothing step as a layer in these networks able to adaptively modulate the degree of smoothing required by each brain volume to better accomplish a given data analysis task. The viability is evaluated on real fMRI data where subjects did alternate between left and right finger tapping tasks.
Original languageEnglish
Title of host publicationProceedings of the 2017 International Workshop on Pattern Recognition in Neuroimaging
Number of pages4
PublisherIEEE
Publication date2017
DOIs
StatePublished - 2017
Event2017 International Workshop on Pattern Recognition in Neuroimaging - Toronto, Canada

Conference

Conference2017 International Workshop on Pattern Recognition in Neuroimaging
LocationUniversity of Toronto
CountryCanada
CityToronto
Period21/06/201723/06/2017

 

Intrinsic Grassmann Averages for Online Linear and Robust Subspace Learning

År: 2017

Intrinsic Grassmann Averages for Online Linear and Robust Subspace Learning

Chakraborty, R., Hauberg, S. & Vemuri, B. C. 2017 Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition. IEEE, p. 801-809

Publication: ResearchArticle in proceedings – Annual report year: 2018

Principal Component Analysis (PCA) is a fundamental method for estimating a linear subspace approximation to high-dimensional data. Many algorithms exist in literature to achieve a statistically robust version of PCA called RPCA. In this paper, we present a geometric framework for computing the principal linear subspaces in both situations that amounts to computing the intrinsic average on the space of all subspaces (the Grassmann manifold). Points on this manifold are defined as the subspaces spanned by K-tuples of observations. We show that the intrinsic Grassmann average of these subspaces coincide with the principal components of the observations when they are drawn from a Gaussian distribution. Similar results are also shown to hold for the RPCA. Further, we propose an efficient online algorithm to do subspace averaging which is of linear complexity in terms of number of samples and has a linear convergence rate. When the data has outliers, our proposed online robust subspace averaging algorithm shows significant performance (accuracy and computation time) gain over a recently published RPCA methods with publicly accessible code. We have demonstrated competitive performance of our proposed online subspace algorithm method on one synthetic and two real data sets. Experimental results depicting stability of our proposed method are also presented. Furthermore, on two real outlier corrupted datasets, we present comparison experiments showing lower reconstruction error using our online RPCA algorithm. In terms of reconstruction error and time required, both our algorithms outperform the competition.
Original languageEnglish
Title of host publicationProceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition
PublisherIEEE
Publication date2017
Pages801-809
DOIs
StatePublished - 2017
Event2017 IEEE Conference on Computer Vision and Pattern Recognition - Honolulu, United States

Conference

Conference2017 IEEE Conference on Computer Vision and Pattern Recognition
CountryUnited States
CityHonolulu
Period21/07/201726/07/2017

 

CloudScan - A Configuration-Free Invoice Analysis System Using Recurrent Neural Networks

År: 2017

CloudScan - A Configuration-Free Invoice Analysis System Using Recurrent Neural Networks

Palm, R. B., Winther, O. & Laws, F. 2017 Proceedings of 2017 14th IAPR International Conference on Document Analysis and Recognition . IEEE, p. 406-13 (2017 14th Iapr International Conference on Document Analysis and Recognition (icdar)).

Publication: Research - peer-reviewArticle in proceedings – Annual report year: 2018

We present CloudScan; an invoice analysis system that requires zero configuration or upfront annotation. In contrast to previous work, CloudScan does not rely on templates of invoice layout, instead it learns a single global model of invoices that naturally generalizes to unseen invoice layouts. The model is trained using data automatically extracted from end-user provided feedback. This automatic training data extraction removes the requirement for users to annotate the data precisely. We describe a recurrent neural network model that can capture long range context and compare it to a baseline logistic regression model corresponding to the current CloudScan production system. We train and evaluate the system on 8 important fields using a dataset of 326,471 invoices. The recurrent neural network and baseline model achieve 0.891 and 0.887 average F1 scores respectively on seen invoice layouts. For the harder task of unseen invoice layouts, the recurrent neural network model outperforms the baseline with 0.840 average F1 compared to 0.788.
Original languageEnglish
Title of host publicationProceedings of 2017 14th IAPR International Conference on Document Analysis and Recognition
PublisherIEEE
Publication date2017
Pages406-13
ISBN (print)9781538635858
DOIs
StatePublished - 2017
Event2017 14th IAPR International Conference on Document Analysis and Recognition
- Kyoto, Japan

Conference

Conference2017 14th IAPR International Conference on Document Analysis and Recognition
LocationKyoto Terrsa
CountryJapan
CityKyoto
Period13/11/201715/11/2017
Series2017 14th Iapr International Conference on Document Analysis and Recognition (icdar)
ISSN2379-2140

 

Bayesian inference for spatio-temporal spike-and-slab priors

År: 2017

Bayesian inference for spatio-temporal spike-and-slab priors

Andersen, M. R., Vehtari, A., Winther, O. & Hansen, L. K. 2017 In : Journal of Machine Learning Research. 18

Publication: Research - peer-reviewJournal article – Annual report year: 2018

In this work, we address the problem of solving a series of underdetermined linear inverse problemblems subject to a sparsity constraint. We generalize the spike-and-slab prior distribution to encode a priori correlation of the support of the solution in both space and time by imposing a transformed Gaussian process on the spike-and-slab probabilities. An expectation propagation (EP) algorithm for posterior inference under the proposed model is derived. For large scale problems, the standard EP algorithm can be prohibitively slow. We therefore introduce three different approximation schemes to reduce the computational complexity. Finally, we demonstrate the proposed model using numerical experiments based on both synthetic and real data sets.
Original languageEnglish
JournalJournal of Machine Learning Research
Volume18
ISSN1533-7928
StatePublished - 2017

  PDF

Reducing the rate and duration of Re- ADMISsions among patients with unipolar disorder and bipolar disorder using smartphone-based monitoring and treatment - the RADMIS trials: Study protocol for two randomized controlled trials

År: 2017

Reducing the rate and duration of Re- ADMISsions among patients with unipolar disorder and bipolar disorder using smartphone-based monitoring and treatment - the RADMIS trials: Study protocol for two randomized controlled trials

Faurholt-Jepsen, M., Frost, M., Martiny, K., Tuxen, N., Rosenberg, N., Busk, J., Winther, O., Bardram, J. E. & Kessing, L. V. 2017 In : Trials. 18, 1, 13 p., 277

Publication: Research - peer-reviewJournal article – Annual report year: 2017

Background: Unipolar and bipolar disorder combined account for nearly half of all morbidity and mortality due to mental and substance use disorders, and burden society with the highest health care costs of all psychiatric and neurological disorders. Among these, costs due to psychiatric hospitalization are a major burden. Smartphones comprise an innovative and unique platform for the monitoring and treatment of depression and mania. No prior trial has investigated whether the use of a smartphone-based system can prevent re-admission among patients discharged from hospital. The present RADMIS trials aim to investigate whether using a smartphone-based monitoring and treatment system, including an integrated clinical feedback loop, reduces the rate and duration of re-admissions more than standard treatment in unipolar disorder and bipolar disorder. Methods: The RADMIS trials use a randomized controlled, single-blind, parallel-group design. Patients with unipolar disorder and patients with bipolar disorder are invited to participate in each trial when discharged from psychiatric hospitals in The Capital Region of Denmark following an affective episode and randomized to either (1) a smartphonebased monitoring system including (a) an integrated feedback loop between patients and clinicians and (b) context-aware cognitive behavioral therapy (CBT) modules (intervention group) or (2) standard treatment (control group) for a 6-month trial period. The trial started in May 2017. The outcomes are (1) number and duration of re-admissions (primary), (2) severity of depressive and manic (only for patients with bipolar disorder) symptoms; psychosocial functioning; number of affective episodes (secondary), and (3) perceived stress, quality of life, self-rated depressive symptoms, self-rated manic symptoms (only for patients with bipolar disorder), recovery, empowerment, adherence to medication, wellbeing, ruminations, worrying, and satisfaction (tertiary). A total of 400 patients (200 patients with unipolar disorder and 200 patients with bipolar disorder) will be included in the RADMIS trials. Discussion: If the smartphone-based monitoring system proves effective in reducing the rate and duration of readmissions, there will be basis for using a system of this kind in the treatment of unipolar and bipolar disorder in general and on a larger scale.

Original languageEnglish
Article number277
JournalTrials
Volume18
Issue number1
Number of pages13
ISSN1745-6215
DOIs
StatePublished - 2017

 

Can smartphone-based electronic markers discriminate between patients with bipolar disorder, healthy first-degree relatives and healthy control individuals

År: 2017

Can smartphone-based electronic markers discriminate between patients with bipolar disorder, healthy first-degree relatives and healthy control individuals

Stanislaus, S., Faurholt-Jepsen, M., Vinberg, M., Winther, O., Frost, M. G., Bardram, J. E. & Kessing, L. 2017 In : Bipolar Disorders (English Edition, Online). 19, p. 128-128 1 p.

Publication: Research - peer-reviewConference abstract in journal – Annual report year: 2017

Original languageEnglish
JournalBipolar Disorders (English Edition, Online)
Volume19
Pages (from-to)128-128
Number of pages1
ISSN1399-5618
StatePublished - 2017
Event19th Annual Conference of the International Society for Bipolar Disorders - Washington DC, United States

Conference

Conference19th Annual Conference of the International Society for Bipolar Disorders
Number19
CountryUnited States
CityWashington DC
Period04/05/201707/05/2017

 

Separable explanations of neural network decisions

År: 2017

Separable explanations of neural network decisions

Rieger, L. 2017 Proceedings of the 31st Conference on Neural Information Processing Systems (NIPS 2017). 8 p.

Publication: Research - peer-reviewArticle in proceedings – Annual report year: 2017

Deep Taylor Decomposition is a method used to explain neural network decisions.
When applying this method to non-dominant classifications, the resulting explanation does not reflect important features for the chosen classification. We propose that this is caused by the dense layers and propose a method to alleviate the effect by applying regularization. We assess the result by measuring the quality of the resulting explanations objectively and subjectively.
Original languageEnglish
Title of host publicationProceedings of the 31st Conference on Neural Information Processing Systems (NIPS 2017)
Number of pages8
Publication date2017
StatePublished - 2017
Event31st Conference on Neural Information Processing Systems - Long Beach, United States

Conference

Conference31st Conference on Neural Information Processing Systems
CountryUnited States
CityLong Beach
Period04/12/201709/12/2017

 

Modeling the Temporal Nature of Human Behavior for Demographics Prediction

År: 2017

Modeling the Temporal Nature of Human Behavior for Demographics Prediction

Felbo, B., Sundsøy, P., Pentland, A., Jørgensen, S. L. & Montjoye, Y-A. 2017 Machine Learning and Knowledge Discovery in Databases. Altun, Y., Das, . K., Mielikäinen, T., Malerba, D., Stefanowski, J., Read, J., Žitnik, M., Ceci, M. & Džeroski, S. (eds.). Springer, Vol. 10536, p. 140-152 13 p. (Lecture Notes in Computer Science).

Publication: Research - peer-reviewBook chapter – Annual report year: 2017

Mobile phone metadata is increasingly used for humanitarian purposes in developing countries as traditional data is scarce. Basic demographic information is however often absent from mobile phone datasets, limiting the operational impact of the datasets. For these reasons, there has been a growing interest in predicting demographic information from mobile phone metadata. Previous work focused on creating increasingly advanced features to be modeled with standard machine learning algorithms. We here instead model the raw mobile phone metadata directly using deep learning, exploiting the temporal nature of the patterns in the data. From high-level assumptions we design a data representation and convolutional network architecture for modeling patterns within a week. We then examine three strategies for aggregating patterns across weeks and show that our method reaches state-of-the-art accuracy on both age and gender prediction using only the temporal modality in mobile metadata. We finally validate our method on low activity users and evaluate the modeling assumptions.
Original languageEnglish
Title of host publicationMachine Learning and Knowledge Discovery in Databases
EditorsYasemin Altun, Kamalika Das, Taneli Mielikäinen, Donato Malerba, Jerzy Stefanowski, Jesse Read, Marinka Žitnik, Michelangelo Ceci, Sašo Džeroski
Number of pages13
Volume10536
PublisherSpringer
Publication date2017
Pages140-152
ISBN (print)978-3-319-71272-7
ISBN (electronic)978-3-319-71273-4
DOIs
StatePublished - 2017
EventThe European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases 2017 - Skopje, Macedonia, The Former Yugoslav Republic of

Conference

ConferenceThe European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases 2017
CountryMacedonia, The Former Yugoslav Republic of
CitySkopje
Period18/09/201722/09/2017
SeriesLecture Notes in Computer Science
ISSN0302-9743

  PDF

The role of gender in social network organization

År: 2017

The role of gender in social network organization

Psylla, I., Sapiezynski, P., Mones, E. & Jørgensen, S. L. 2017 In : P L o S One. 12, 12, 21 p., e0189873

Publication: Research - peer-reviewJournal article – Annual report year: 2017

The digital traces we leave behind when engaging with the modern world offer an interesting lens through which we study behavioral patterns as expression of gender. Although gender differentiation has been observed in a number of settings, the majority of studies focus on a single data stream in isolation. Here we use a dataset of high resolution data collected using mobile phones, as well as detailed questionnaires, to study gender differences in a large cohort. We consider mobility behavior and individual personality traits among a group of more than 800 university students. We also investigate interactions among them expressed via person-to-person contacts, interactions on online social networks, and telecommunication. Thus, we are able to study the differences between male and female behavior captured through a multitude of channels for a single cohort. We find that while the two genders are similar in a number of aspects, there are robust deviations that include multiple facets of social interactions, suggesting the existence of inherent behavioral differences. Finally, we quantify how aspects of an individual's characteristics and social behavior reveals their gender by posing it as a classification problem. We ask: How well can we distinguish between male and female study participants based on behavior alone? Which behavioral features are most predictive?
Original languageEnglish
Article numbere0189873
JournalP L o S One
Volume12
Issue number12
Number of pages21
ISSN1932-6203
DOIs
StatePublished - 2017

Bibliographical note

Copyright: © 2017 Psylla et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

  PDF

Correlations between human mobility and social interaction reveal general activity patterns

År: 2017

Correlations between human mobility and social interaction reveal general activity patterns

Mollgaard, A., Jørgensen, S. L. & Mathiesen, J. 2017 In : P L o S One. 12, 12, 16 p., e0188973

Publication: Research - peer-reviewJournal article – Annual report year: 2017

A day in the life of a person involves a broad range of activities which are common across many people. Going beyond diurnal cycles, a central question is: to what extent do individuals act according to patterns shared across an entire population? Here we investigate the interplay between different activity types, namely communication, motion, and physical proximity by analyzing data collected from smartphones distributed among 638 individuals. We explore two central questions: Which underlying principles govern the formation of the activity patterns? Are the patterns specific to each individual or shared across the entire population? We find that statistics of the entire population allows us to successfully predict 71% of the activity and 85% of the inactivity involved in communication, mobility, and physical proximity. Surprisingly, individual level statistics only result in marginally better predictions, indicating that a majority of activity patterns are shared across our sample population. Finally, we predict short-term activity patterns using a generalized linear model, which suggests that a simple linear description might be sufficient to explain a wide range of actions, whether they be of social or of physical character.
Original languageEnglish
Article numbere0188973
JournalP L o S One
Volume12
Issue number12
Number of pages16
ISSN1932-6203
DOIs
StatePublished - 2017

Bibliographical note

Copyright: © 2017 Mollgaard et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

  PDF

Boganmeldelse: Pascal - Matematik og mirakler

År: 2017

Boganmeldelse: Pascal - Matematik og mirakler

Hansen, V. L. 2017 In : Aktuel Naturvidenskab. 2017, 6, p. 44 1 p.

Publication: CommunicationLiterature review – Annual report year: 2017

Original languageEnglish
JournalAktuel Naturvidenskab
Volume2017
Issue number6
Pages (from-to)44
Number of pages1
ISSN1399-2309
StatePublished - 2017

Bibliographical note

Anmeldelse af bogen "Pascal - Matematik og mirakler" skrevet af Carl Henrik Koch.

  PDF

Risk for affective disorders is associated with greater prefrontal gray matter volumes: A prospective longitudinal study

År: 2017

Risk for affective disorders is associated with greater prefrontal gray matter volumes: A prospective longitudinal study

Macoveanu, J., Baaré, W., Madsen, K. H., Kessing, L. V., Siebner, H. R. & Vinberg, M. 2017 In : NeuroImage: Clinical. 27 p.

Publication: Research - peer-reviewJournal article – Annual report year: 2017

Background: Major depression and bipolar disorders aggregates in families and are linked with a wide range of neurobiological abnormalities including cortical gray matter (GM) alterations. Prospective studies of individuals at familial risk may expose the neural mechanisms underlying risk transmission. Methods: We used voxel based morphometry to investigate changes in regional GM brain volume, over a seven-year period, in 37 initially healthy individuals having a mono- or di-zygotic twin diagnosed with major depression or bipolar disorder (high-risk group; mean age 41.6 yrs.) as compared to 36 individuals with no history of affective disorders in the index twin and firstdegree relatives (low-risk group; mean age 38.5 yrs.). Results: Groups did not differ in regional GM volume changes over time. However, independent of time, high-risk twins had significantly greater GM volumes in bilateral dorsal anterior cingulate, inferior frontal gyrus and temporoparietal regions as compared to low-risk twins. Further, individuals who developed an affective disorder at follow-up (n=12), had relatively the largest GM volumes, both at baseline and follow-up, in the right dorsal anterior cingulate cortex and right inferior frontal cortex compared to high- and low-risk twins who remained well at follow-up. Conclusion: This pattern of apparently stable grater regional GM volume may constitute a neural marker of an increased risk for developing an affective disorder in individuals at familial risk.
Original languageEnglish
JournalNeuroImage: Clinical
Number of pages27
ISSN2213-1582
DOIs
StatePublished - 2017

  PDF

Scholia and scientometrics with Wikidata

År: 2017

Scholia and scientometrics with Wikidata

Nielsen, F. Å., Mietchen, D. & Willighagen, E. 2017 Joint Proceedings of the 1st International Workshop on Scientometrics and 1st International Workshop on Enabling Decentralised Scholarly Communication. 16 p.

Publication: Research - peer-reviewArticle in proceedings – Annual report year: 2017

Scholia is a tool to handle scientific bibliographic information through Wikidata. The Scholia Web service creates on-the-fly scholarly profiles for researchers, organizations, journals, publishers, individual scholarly works, and for research topics. To collect the data, it queries the SPARQL-based Wikidata Query Service. Among several display formats available in Scholia are lists of publications for individual researchers and organizations, publications per year, employment timelines, as well as coauthor and topic networks and citation graphs. The Python package implementing the Web service is also able to format Wikidata bibliographic entries for use in LaTeX/BIBTeX.
Original languageEnglish
Title of host publicationJoint Proceedings of the 1st International Workshop on Scientometrics and 1st International Workshop on Enabling Decentralised Scholarly Communication
Number of pages16
Publication date2017
DOIs
StatePublished - 2017
EventScientometrics and Enabling Decentralised Scholarly Communication - Portorož, Slovenia

Conference

ConferenceScientometrics and Enabling Decentralised Scholarly Communication
CountrySlovenia
CityPortorož
Period28/05/201728/05/2017

 

Fostering Bilateral Patient-Clinician Engagement in Active Self-Tracking of Subjective Experience

År: 2017

Fostering Bilateral Patient-Clinician Engagement in Active Self-Tracking of Subjective Experience

Larsen, J. E., Christiansen, T. B. & Eskelund, K. 2017 Proceedings of Pervasive Health 2017. Association for Computing Machinery, 4 p.

Publication: Research - peer-reviewArticle in proceedings – Annual report year: 2017

In this position paper we describe select aspects of our experience with health-related self-tracking, the data generated, and processes surrounding those. In
particular we focus on how bilateral patient-clinician engagement may be fostered by the combination of technology and method. We exemplify with a case
study where a PTSD-suffering veteran has been selftracking a specific symptom precursor. The availability of high-resolution self-tracking data on the occurrences
of even a single symptom created new opportunities in the therapeutic process for identifying underlying triggers of symptoms. The patient was highly engaged
in self-tracking and sharing the collected data. We suggest a key reason was the collaborative effort in defining the data collection protocol and discussion of
the data. The therapist also engaged highly in the selftracking data, as it supported the existing therapeutic process in reaching insights otherwise unobtainable.
Original languageEnglish
Title of host publicationProceedings of Pervasive Health 2017
Number of pages4
PublisherAssociation for Computing Machinery
Publication date2017
DOIs
StatePublished - 2017
EventPervasive Health 2017 - Barcelona, Spain

Conference

ConferencePervasive Health 2017
CountrySpain
CityBarcelona
Period23/05/201726/05/2017

 

New frontiers of quantified self 3: Exploring understudied categories of users

År: 2017

New frontiers of quantified self 3: Exploring understudied categories of users

Rapp, A., Cena, F., Kay, J., Kummerfeld, B., Hopfgartner, F., Plumbaum, T., Larsen, J. E., Epstein, D. A. & Gouveia, R. 2017 UbiComp/ISWC 2017 - Adjunct Proceedings of the 2017 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2017 ACM International Symposium on Wearable Computers. Association for Computing Machinery, p. 861-864 4 p.

Publication: Research - peer-reviewArticle in proceedings – Annual report year: 2017

Quantified Self (QS) field needs to start thinking of how situated needs may affect the use of self-tracking technologies. In this workshop we will focus on the idiosyncrasies of specific categories of users.

Original languageEnglish
Title of host publicationUbiComp/ISWC 2017 - Adjunct Proceedings of the 2017 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2017 ACM International Symposium on Wearable Computers
Number of pages4
PublisherAssociation for Computing Machinery
Publication date2017
Pages861-864
ISBN (electronic)9781450351904
DOIs
StatePublished - 2017
Event2017 ACM International Joint Conference on Pervasive and Ubiquitous Computing and ACM International Symposium on Wearable Computers, UbiComp/ISWC 2017 - Maui, United States

Conference

Conference2017 ACM International Joint Conference on Pervasive and Ubiquitous Computing and ACM International Symposium on Wearable Computers, UbiComp/ISWC 2017
CountryUnited States
CityMaui
Period11/09/201715/09/2017
SponsorACM SIGCHI, et al., FlyBits, Intel, National Science Foundation, SIGMOBILE, ACM SIGCHI, et al., FlyBits, Intel, National Science Foundation, SIGMOBILE

  PDF

Active Self-Tracking of Subjective Experience with a One-Button Wearable: A Case Study in Military PTSD

År: 2017

Active Self-Tracking of Subjective Experience with a One-Button Wearable: A Case Study in Military PTSD

Larsen, J. E., Eskelund, K. & Christiansen, T. B. 2017 Proceedings of the 2nd Computing and Mental Health workshop at ACM CHI 2017.. 5 p.

Publication: Research - peer-reviewArticle in proceedings – Annual report year: 2017

We describe a case study with the participation of a Danish veteran suffering from post-traumatic stress disorder (PTSD). As part of psychotherapeutic treatment the participant and therapist have used our novel technique for instrumenting self-tracking of select aspects of subjective experience using a one-button wearable device. The instrumentation system is described along with the specific self-tracking protocol which defined the participant’s self-tracking of a single symptom, namely the occurrences of a bodily experienced precursor to hyperarousal. Results from the case study demonstrate how self-tracking data on a single symptom
collected by a patient can provide valuable input to the therapeutic process. Specifically, it facilitated identification of crucial details otherwise unavailable from the clinical assessment and even became decisive in disentangling different symptoms and their causes.
Original languageEnglish
Title of host publicationProceedings of the 2nd Computing and Mental Health workshop at ACM CHI 2017.
Number of pages5
Publication date2017
StatePublished - 2017
Event2nd Computing and Mental Health workshop at ACM CHI 2017. - Denver, United States

Workshop

Workshop2nd Computing and Mental Health workshop at ACM CHI 2017.
CountryUnited States
CityDenver
Period06/05/201711/05/2017

 

Open semantic analysis: The case of word level semantics in Danish

År: 2017

Open semantic analysis: The case of word level semantics in Danish

Nielsen, F. Å. & Hansen, L. K. 2017 Proceedings of 8th Language and Technology Conference . 5 p.

Publication: Research - peer-reviewArticle in proceedings – Annual report year: 2017

The present research is motivated by the need for accessible and efficient tools for automated semantic analysis in Danish. We are interested in tools that are completely open, so they can be used by a critical public, in public administration, non-governmental organizations and businesses. We describe data-driven models for Danish semantic relatedness, word intrusion and sentiment prediction. Open Danish corpora were assembled and unsupervised learning implemented for explicit semantic analysis and with Gensim’s Word2vec model. We evaluate the performance of the two models on three different annotated word datasets. We test the semantic representations’ alignment with single word sentiment using supervised learning. We find that logistic regression and large random forests perform well with Word2vec features.
Original languageEnglish
Title of host publicationProceedings of 8th Language and Technology Conference
Number of pages5
Publication date2017
StatePublished - 2017
Event8th Language and Technology Conference - Poznan, Poland

Conference

Conference8th Language and Technology Conference
CountryPoland
CityPoznan
Period17/11/201719/11/2017

 

Difference-of-Convex optimization for variational kl-corrected inference in dirichlet process mixtures

År: 2017

Difference-of-Convex optimization for variational kl-corrected inference in dirichlet process mixtures

Bonnevie, R., Schmidt, M. N. & Mørup, M. 2017 Proceedings of 2017 IEEE 27th International Workshop on Machine Learning for Signal Processing (MLSP). IEEE, p. 1-6 6 p.

Publication: Research - peer-reviewArticle in proceedings – Annual report year: 2017

Variational methods for approximate inference in Bayesian models optimise a lower bound on the marginal likelihood, but the optimization problem often suffers from being nonconvex and high-dimensional. This can be alleviated by working in a collapsed domain where a part of the parameter space is marginalized. We consider the KL-corrected collapsed variational bound and apply it to Dirichlet process mixture models, allowing us to reduce the optimization space considerably. We find that the variational bound exhibits consistent and exploitable structure, allowing the application of difference-of-convex optimization algorithms. We show how this yields an interpretable fixed-point update algorithm in the collapsed setting for the Dirichlet process mixture model. We connect this update formula to classical coordinate ascent updates, illustrating that the proposed improvement surprisingly reduces to the traditional scheme.
Original languageEnglish
Title of host publicationProceedings of 2017 IEEE 27th International Workshop on Machine Learning for Signal Processing (MLSP)
Number of pages6
PublisherIEEE
Publication date2017
Pages1-6
ISBN (print)978-1-5090-6342-0
ISBN (electronic)978-1-5090-6341-3
DOIs
StatePublished - 2017
Event2017 IEEE international workshop on machine learning for signal processing - Tokyo, Jamaica

Workshop

Workshop2017 IEEE international workshop on machine learning for signal processing
Number27
LocationRoppongi
CountryJamaica
CityTokyo
Period25/09/201728/09/2017
Internet address

 

A Disentangled Recognition and Nonlinear Dynamics Model for Unsupervised Learning

År: 2017

A Disentangled Recognition and Nonlinear Dynamics Model for Unsupervised Learning

Fraccaro, M., Kamronn, S. D., Paquet, U. & Winther, O. 2017 Proceedings of 31st Conference on Neural Information Processing Systems . 13 p.

Publication: Research - peer-reviewArticle in proceedings – Annual report year: 2017

This paper takes a step towards temporal reasoning in a dynamically changing video, not in the pixel space that constitutes its frames, but in a latent space that describes the non-linear dynamics of the objects in its world. We introduce the Kalman variational auto-encoder, a framework for unsupervised learning of sequential data that disentangles two latent representations: an object’s representation, coming from a recognition model, and a latent state describing its dynamics. As a result, the evolution of the world can be imagined and missing data imputed, both without the need to generate high dimensional frames at each time step. The model is trained end-to-end on videos of a variety of simulated physical systems, and outperforms competing methods in generative and missing data imputation tasks.
Original languageEnglish
Title of host publicationProceedings of 31st Conference on Neural Information Processing Systems
Number of pages13
Publication date2017
StatePublished - 2017
Event31st Conference on Neural Information Processing Systems - Long Beach, United States

Conference

Conference31st Conference on Neural Information Processing Systems
CountryUnited States
CityLong Beach
Period04/12/201709/12/2017

  PDF

End-to-end information extraction without token-level supervision

År: 2017

End-to-end information extraction without token-level supervision

Palm, R. B., Hovy, D., Laws, F. & Winther, O. 2017 In : ArXiv. 2017, 5 p., 1707.04913

Publication: Research - peer-reviewJournal article – Annual report year: 2017

Most state-of-the-art information extraction approaches rely on token-level labels to find the areas of interest in text. Unfortunately, these labels are time-consuming and costly to create, and consequently, not available for many real-life IE tasks. To make matters worse, token-level labels are usually not the desired output, but just an intermediary step. End-to-end (E2E) models, which take raw text as input and produce the desired output directly, need not depend on token-level labels. We propose an E2E model based on pointer networks, which can be trained directly on pairs of raw input and output text. We evaluate our model on the ATIS data set, MIT restaurant corpus and the MIT movie corpus and compare to neural baselines that do use token-level labels. We achieve competitive results, within a few percentage points of the baselines, showing the feasibility of E2E information extraction without the need for token-level labels. This opens up new possibilities, as for many tasks currently addressed by human extractors, raw input and output data are available, but not token-level labels.
Original languageEnglish
Article number1707.04913
JournalArXiv
Volume2017
Number of pages5
StatePublished - 2017

  PDF

Sequence Classification Using Third-Order Moments

År: 2017

Sequence Classification Using Third-Order Moments

Troelsgaard, R. & Hansen, L. K. 2017 In : Neural Computation. 30, 1, p. 216-236

Publication: Research - peer-reviewJournal article – Annual report year: 2018

Model-based classification of sequence data using a set of hidden Markov models is a well-known technique. The involved score function, which is often based on the class-conditional likelihood, can, however, be computationally demanding, especially for long data sequences. Inspired by recent theoretical advances in spectral learning of hidden Markov models, we propose a score function based on third-order moments. In particular, we propose to use the Kullback-Leibler divergence between theoretical and empirical third-order moments for classification of sequence data with discrete observations. The proposed method provides lower computational complexity at classification time than the usual likelihood-based methods. In order to demonstrate the properties of the proposed method, we perform classification of both simulated data and empirical data from a human activity recognition study.
Original languageEnglish
JournalNeural Computation
Volume30
Issue number1
Pages (from-to)216-236
ISSN0899-7667
DOIs
StatePublished - 2017

  PDF

Class attendance, peer similarity, and academic performance in a large field study

År: 2017

Class attendance, peer similarity, and academic performance in a large field study

Kassarnig, V., Bjerre-Nielsen, A., Mones, E., Lehmann, S. & Lassen, D. D. 2017 In : P L o S One. 12, 11, 15 p., e0187078

Publication: Research - peer-reviewJournal article – Annual report year: 2017

Identifying the factors that determine academic performance is an essential part of educational research. Existing research indicates that class attendance is a useful predictor of subsequent course achievements. The majority of the literature is, however, based on surveys and self-reports, methods which have well-known systematic biases that lead to limitations on conclusions and generalizability as well as being costly to implement. Here we propose a novel method for measuring class attendance that overcomes these limitations by using location and bluetooth data collected from smartphone sensors. Based on measured attendance data of nearly 1,000 undergraduate students, we demonstrate that early and consistent class attendance strongly correlates with academic performance. In addition, our novel dataset allows us to determine that attendance among social peers was substantially correlated (>0.5), suggesting either an important peer effect or homophily with respect to attendance.
Original languageEnglish
Article numbere0187078
JournalP L o S One
Volume12
Issue number11
Number of pages15
ISSN1932-6203
DOIs
StatePublished - 2017

Bibliographical note

© 2017 Kassarnig et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

  PDF

Evidence of complex contagion of information in social media: An experiment using Twitter bots

År: 2017

Evidence of complex contagion of information in social media: An experiment using Twitter bots

Mønsted, B. M., Sapiezynski, P., Ferrara, E. & Jørgensen, S. L. 2017 In : P L o S One. 12, 9, 12 p., e0184148

Publication: Research - peer-reviewJournal article – Annual report year: 2017

It has recently become possible to study the dynamics of information diffusion in techno-social systems at scale, due to the emergence of online platforms, such as Twitter, with millions of users. One question that systematically recurs is whether information spreads according to simple or complex dynamics: does each exposure to a piece of information have an independent probability of a user adopting it (simple contagion), or does this probability depend instead on the number of sources of exposure, increasing above some threshold (complex contagion)? Most studies to date are observational and, therefore, unable to disentangle the effects of confounding factors such as social reinforcement, homophily, limited attention, or network community structure. Here we describe a novel controlled experiment that we performed on Twitter using 'social bots' deployed to carry out coordinated attempts at spreading information. We propose two Bayesian statistical models describing simple and complex contagion dynamics, and test the competing hypotheses. We provide experimental evidence that the complex contagion model describes the observed information diffusion behavior more accurately than simple contagion. Future applications of our results include more effective defenses against malicious propaganda campaigns on social media, improved marketing and advertisement strategies, and design of effective network intervention techniques.
Original languageEnglish
Article numbere0184148
JournalP L o S One
Volume12
Issue number9
Number of pages12
ISSN1932-6203
DOIs
StatePublished - 2017

Bibliographical note

Copyright: © 2017 Mønsted et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

 

Unraveling fermentation data – a Novozymes case study

År: 2017

Unraveling fermentation data – a Novozymes case study

Baum, A., Vermue, L., Moiseyenko, R., Jørgensen, T. M. & Devantier, R. 2017

Publication: Research - peer-reviewConference abstract for conference – Annual report year: 2017

Industrial fermentation processes are monitored using a variety of sensors. Typically, measurements are taken through-out the entire production process. Production may be carried out under supervision of different operators (operator variation), on different sites (global variation), in different buildings and/or in different tanks (local variation). However, up to now processes are mainly controlled according to traditional recipes and experience
Original languageEnglish
Publication date2017
Number of pages1
StatePublished - 2017
EventRecent Advances in Fermentation Technology (RAFT 2017) - Florida, United States

Conference

ConferenceRecent Advances in Fermentation Technology (RAFT 2017)
LocationHyatt Coconut Point
CountryUnited States
CityFlorida
Period29/10/201701/11/2017

 

Scholia, Scientometrics and Wikidata

År: 2017

Scholia, Scientometrics and Wikidata

Nielsen, F., Mietchen, D. & Willighagen, E. 2017 The Semantic Web: ESWC 2017 Satellite Events. Springer, Vol. 10577, p. 237-259 (Lecture Notes in Computer Science, Vol. 10577).

Publication: Research - peer-reviewArticle in proceedings – Annual report year: 2017

Scholia is a tool to handle scientific bibliographic information through Wikidata. The Scholia Web service creates on-the-fly scholarly profiles for researchers, organizations, journals, publishers, individual scholarly works, and for research topics. To collect the data, it queries the SPARQL-based Wikidata Query Service. Among several display formats available in Scholia are lists of publications for individual researchers and organizations, plots of publications per year, employment timelines, as well as co-author and topic networks and citation graphs. The Python package implementing the Web service is also able to format Wikidata bibliographic entries for use in LaTeX/BIBTeX. Apart from detailing Scholia, we describe how Wikidata has been used for bibliographic information and we also provide some scientometric statistics on this information.
Original languageEnglish
Title of host publicationThe Semantic Web: ESWC 2017 Satellite Events
Volume10577
PublisherSpringer
Publication date2017
Pages237-259
DOIs
StatePublished - 2017
Event14th ESCW - European Semantic Web Conference 2017 - Portoroz, Slovenia

Conference

Conference14th ESCW - European Semantic Web Conference 2017
Number14
CountrySlovenia
CityPortoroz
Period28/05/201701/06/2017
Internet address
SeriesLecture Notes in Computer Science
Volume10577
ISSN0302-9743

 

Why Do We Fall into Sync with Others? Interpersonal Synchronization and the Brain's Optimization Principle

År: 2017

Why Do We Fall into Sync with Others? Interpersonal Synchronization and the Brain's Optimization Principle

Koban, L., Ramamoorthy, A. & Konvalinka, I. 2017 In : Social Neuroscience. 10 p.

Publication: Research - peer-reviewJournal article – Annual report year: 2017

Spontaneous interpersonal synchronization of rhythmic behavior such as gait or clapping is a ubiquitous phenomenon in human interactions, and is potentially important for social relationships and action understanding. Although several authors have suggested a role of the mirror neuron system in interpersonal coupling, the underlying brain mechanisms are not well understood. Here we argue that more general theories of neural computations, namely predictive coding and the Free Energy Principle, could explain interpersonal coordination dynamics. Each brain minimizes coding costs by reducing the mismatch between the representations of observed and own motor behavior. Continuous mutual prediction and alignment result in an overall minimization of free energy, thus forming a stable attractor state.
Original languageEnglish
JournalSocial Neuroscience
Number of pages10
ISSN1747-0919
DOIs
StateAccepted/In press - 2017

  PDF

Transformations Based on Continuous Piecewise-Affine Velocity Fields

År: 2017

Transformations Based on Continuous Piecewise-Affine Velocity Fields

Freifeld, O., Hauberg, S., Batmanghelich, K. & Fisher, J. W. 2017 In : I E E E Transactions on Pattern Analysis and Machine Intelligence. 39, 12, p. 2496-2509

Publication: Research - peer-reviewJournal article – Annual report year: 2017

We propose novel finite-dimensional spaces of well-behaved transformations. The latter are obtained by (fast and highly-accurate) integration of continuous piecewise-affine velocity fields. The proposed method is simple yet highly expressive, effortlessly handles optional constraints (e.g., volume preservation and/or boundary conditions), and supports convenient modeling choices such as smoothing priors and coarse-to-fine analysis. Importantly, the proposed approach, partly due to its rapid likelihood evaluations and partly due to its other properties, facilitates tractable inference over rich transformation spaces, including using Markov-Chain Monte-Carlo methods. Its applications include, but are not limited to: monotonic regression (more generally, optimization over monotonic functions); modeling cumulative distribution functions or histograms; time-warping; image warping; image registration; real-time diffeomorphic image editing; data augmentation for image classifiers. Our GPU-based code is publicly available.
Original languageEnglish
JournalI E E E Transactions on Pattern Analysis and Machine Intelligence
Volume39
Issue number12
Pages (from-to)2496-2509
ISSN0162-8828
DOIs
StatePublished - 2017

 

Using OR + AI to Predict the Optimal Production of Offshore Wind Parks: A Preliminary Study

År: 2017

Using OR + AI to Predict the Optimal Production of Offshore Wind Parks: A Preliminary Study

Fischetti, M. & Fraccaro, M. 2017 Optimization and Decision Science: Methodologies and Applications,. Springer, Vol. 217, p. 203-211 (Optimization and Decision Science: Methodologies and Applications, Vol. 217).

Publication: Research - peer-reviewArticle in proceedings – Annual report year: 2017

In this paper we propose a new use of Machine Learning together with Mathematical Optimization. We investigate the question of whether a machine, trained on a large number of optimized solutions, can accurately estimate the value of the optimized solution for new instances. We focus on instances of a specific problem, namely, the offshore wind farm layout optimization problem. In this problem an offshore site is given, together with the wind statistics and the characteristics of the turbines that need to be built. The optimization wants to determine the optimal allocation of turbines to maximize the park power production, taking the mutual interference between turbines into account. Mixed Integer Programming models and other state-of-the-art optimization techniques, have been developed to solve this problem. Starting with a dataset of 2000+ optimized layouts found by the optimizer, we used supervised learning to estimate the production of new wind parks. Our results show that Machine Learning is able to well estimate the optimal value of offshore wind farm layout problems.
Original languageEnglish
Title of host publicationOptimization and Decision Science: Methodologies and Applications,
Volume217
PublisherSpringer
Publication date2017
Pages203-211
ISBN (print)9783319673073
DOIs
StatePublished - 2017
EventInternational Conference on Optimization and Decision Science -

Conference

ConferenceInternational Conference on Optimization and Decision Science
Period04/09/201707/09/2017
Internet address
SeriesOptimization and Decision Science: Methodologies and Applications
Volume217

 

Deep recurrent conditional random field network for protein secondary prediction

År: 2017

Deep recurrent conditional random field network for protein secondary prediction

Johansen, A. R., Sønderby, S. K., Sønderby, C. K. & Winther, O. 2017 8th ACM International Conference on Bioinformatics, Computational Biology, and Health Informatics. Association for Computing Machinery, p. 73-78 6 p. (Acm-bcb - Proc. Acm Int. Conf. Bioinform., Comput. Biol., Health Informatics).

Publication: Research - peer-reviewArticle in proceedings – Annual report year: 2017

Deep learning has become the state-of-the-art method for predicting protein secondary structure from only its amino acid residues and sequence profile. Building upon these results, we propose to combine a bi-directional recurrent neural network (biRNN) with a conditional random field (CRF), which we call the biRNN-CRF. The biRNN-CRF may be seen as an improved alternative to an autoregressive uni-directional RNN where predictions are performed sequentially conditioning on the prediction in the previous timestep. The CRF is instead nearest neighbor-aware and models for the joint distribution of the labels for all time-steps. We condition the CRF on the output of biRNN, which learns a distributed representation based on the entire sequence. The biRNN-CRF is therefore close to ideally suited for the secondary structure task because a high degree of cross-talk between neighboring elements can be expected. We validate the model on several benchmark datasets. For example, on CB513, a model with 1.7 million parameters, achieves a Q8 accuracy of 69.4 for single model and 70.9 for ensemble, which to our knowledge is state-of-the-art. 1
Original languageEnglish
Title of host publication8th ACM International Conference on Bioinformatics, Computational Biology, and Health Informatics
Number of pages6
PublisherAssociation for Computing Machinery
Publication date2017
Pages73-78
ISBN (print)9781450347228
DOIs
StatePublished - 2017
Event8th ACM International Conference on Bioinformatics, Computational Biology, and Health Informatics - Boston, United States

Conference

Conference8th ACM International Conference on Bioinformatics, Computational Biology, and Health Informatics
CountryUnited States
CityBoston
Period20/08/201723/08/2017
SeriesAcm-bcb - Proc. Acm Int. Conf. Bioinform., Comput. Biol., Health Informatics

  PDF

Electrophysiological evidence for differences between fusion and combination illusions in audiovisual speech perception

År: 2017

Incongruent audiovisual speech stimuli can lead to perceptual illusions such as fusions or combinations. Here, we investigated the underlying audiovisual integration process by measuring ERPs. We observed that visual speech-induced suppression of P2 amplitude (which is generally taken as a measure of audiovisual integration) for fusions was comparable to suppression obtained with fully congruent stimuli, whereas P2 suppression for combinations was larger. We argue that these effects arise because the phonetic incongruency is solved differently for both types of stimuli. This article is protected by copyright. All rights reserved.
Original languageEnglish
JournalEuropean Journal of Neuroscience
ISSN0953-816X
DOIs
StatePublished - 2017

 

A deep learning approach to adherence detection for type 2 diabetics

År: 2017

A deep learning approach to adherence detection for type 2 diabetics

Mohebbi, A., Aradóttir, T. B., Johansen, A. R., Bengtsson, H., Fraccaro, M. & Mørup, M. 2017 Proceedings of 2017 39th Annual International Conference of the Ieee Engineering in Medicine and Biology Society. IEEE, p. 2896-9 (2017 39th Annual International Conference of the Ieee Engineering in Medicine and Biology Society (embc)).

Publication: Research - peer-reviewArticle in proceedings – Annual report year: 2017

Diabetes has become one of the biggest health problems in the world. In this context, adherence to insulin treatment is essential in order to avoid life-threatening complications. In this pilot study, a novel adherence detection algorithm using Deep Learning (DL) approaches was developed for type 2 diabetes (T2D) patients, based on simulated Continuous Glucose Monitoring (CGM) signals. A large and diverse amount of CGM signals were simulated for T2D patients using a T2D adapted version of the Medtronic Virtual Patient (MVP) model for T1D. By using these signals, different classification algorithms were compared using a comprehensive grid search. We contrast a standard logistic regression baseline to Multi- Layer Perceptrons (MLPs) and Convolutional Neural Networks (CNNs). The best classification performance with an average accuracy of 77:5% was achieved with CNN. Hence, this indicates the potential of DL, when considering adherence detection systems for T2D patients.
Original languageEnglish
Title of host publicationProceedings of 2017 39th Annual International Conference of the Ieee Engineering in Medicine and Biology Society
PublisherIEEE
Publication date2017
Pages2896-9
ISBN (print)978-1-5090-2809-2
DOIs
StatePublished - 2017
Event2017 39th Annual International Conference of the Ieee Engineering in Medicine and Biology Society - Jeju Island, Korea, Republic of

Conference

Conference2017 39th Annual International Conference of the Ieee Engineering in Medicine and Biology Society
LocationJUNGMUN Sightseeing Complex
CountryKorea, Republic of
CityJeju Island
Period11/07/201715/07/2017
Series2017 39th Annual International Conference of the Ieee Engineering in Medicine and Biology Society (embc)
ISSN1558-4615

 

From concept to in vivo testing: Microcontainers for oral drug delivery

År: 2017

From concept to in vivo testing: Microcontainers for oral drug delivery

Mazzoni, C., Tentor, F., Andersen, S. S., Nielsen, L. H., Keller, S. S., Alstrøm, T. S., Gundlach, C., Müllertz, A., Marizza, P. & Boisen, A. 2017 In : Journal of Controlled Release. 268, p. 343-351

Publication: Research - peer-reviewJournal article – Annual report year: 2017

This work explores the potential of polymeric micrometer sized devices (microcontainers) as oral drug delivery systems (DDS). Arrays of detachable microcontainers (D-MCs) were fabricated on a sacrificial layer to improve the handling and facilitate the collection of individual D-MCs. A model drug, ketoprofen, was loaded into the microcontainers using supercritical CO2 impregnation, followed by deposition of an enteric coating to protect the drug from the harsh gastric environment and to provide a fast release in the intestine. In vitro, in vivo and ex vivo studies were performed to assess the viability of the D-MCs as oral DDS. D-MCs improved the relative oral bioavailability by 180% within 4h, and increased the absorption rate by 2.4 times compared to the control. This work represents a significant step forward in the translation of these devices from laboratory to clinic.
Original languageEnglish
JournalJournal of Controlled Release
Volume268
Pages (from-to)343-351
ISSN0168-3659
DOIs
StatePublished - 2017

  PDF

Inferring Person-to-person Proximity Using WiFi Signals

År: 2017

Inferring Person-to-person Proximity Using WiFi Signals

Sapiezynski, P., Stopczynski, A., Wind, D. K., Leskovec, J. & Jørgensen, S. L. 2017 Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologie. New York,: Association for Computing Machinery, Vol. 1, 11 p. 24. (Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies ).

Publication: Research - peer-reviewArticle in proceedings – Annual report year: 2017

Today’s societies are enveloped in an ever-growing telecommunication infrastructure. This infrastructure offers important opportunities for sensing and recording a multitude of human behaviors. Human mobility patterns are a prominent example of such a behavior which has been studied based on cell phone towers, Bluetooth beacons, and WiFi networks as proxies for location. However, while mobility is an important aspect of human behavior, understanding complex social systems requires studying not only the movement of individuals, but also their interactions. Sensing social interactions on a large scale is a technical challenge and many commonly used approaches—including RFID badges or Bluetooth scanning—offer only limited scalability. Here we show that it is possible, in a scalable and robust way, to accurately infer person-to-person physical proximity from the lists of WiFi access points measured by smartphones carried by the two individuals. Based on a longitudinal dataset of approximately 800 participants with ground-truth interactions collected over a year, we show that our model performs better than the current state-of-the-art. Our results demonstrate the value of WiFi signals in social sensing as well as potential threats to privacy that they imply.
Original languageEnglish
Title of host publicationProceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologie
Number of pages11
Volume1
Place of PublicationNew York,
PublisherAssociation for Computing Machinery
Publication date2017
Article number24
DOIs
StatePublished - 2017
SeriesProceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies
ISSN2474-9567

 

Optimal allocation of reviewers for peer feedback

År: 2017

Optimal allocation of reviewers for peer feedback

Wind, D. K., Jensen, U. A., Jørgensen, R. M., Hansen, S. L. & Winther, O. 2017 Proceedings of 16th European Conference on eLearning. Academic Conferences and Publishing International

Publication: Research - peer-reviewArticle in proceedings – Annual report year: 2017

Peer feedback is the act of letting students give feedback to each other on submitted work. There are multiple reasons to use peer feedback, including students getting more feedback, time saving for teachers and increased learning by letting students reflect on work by others. In order for peer feedback to be effective students should give and receive useful feedback. A key challenge in peer feedback is allocating the feedback givers in a good way. It is important that reviewers are allocated to submissions such that the feedback distribution is fair - meaning that all students receive good feedback.

In this paper we present a novel way to intelligently allocate reviewers for peer feedback. We train a statistical model to infer the quality of feedback based on a dataset of feedback quality evaluations. This dataset contains more than 20,000 reviews where the receiver of the feedback has indicated the quality of the feedback. Using this model together with historical data we calculate the feedback-giving skill of each student and uses that as input to an allocation algorithm that assigns submissions to reviewers, in order to optimize the feedback quality for all students.

We test the performance of our allocation strategy using real data from over 600 peer feedback sessions and simulate the effects of different allocation strategies. By comparing our method with a random allocation algorithm and a “super-informed oracle” algorithm we demonstrate that we are able to allocate reviewers to submissions in such a way that all submissions receive feedback of similar quality and that we are able to significantly outperform simple random allocation of reviewers. Additionally we investigate the effect of pre-allocating reviews in comparison to allocating reviewers live during the review process and show that live-allocation leads to better results. Our method is robust to reviews not being completed and other real-life quirks and improves as more feedback data is collected.
Original languageEnglish
Title of host publicationProceedings of 16th European Conference on eLearning
PublisherAcademic Conferences and Publishing International
Publication date2017
ISBN (print)978-1-911218-60-9
StatePublished - 2017
Event16th European Conference on eLearning - Porto, Portugal

Conference

Conference16th European Conference on eLearning
CountryPortugal
CityPorto
Period26/10/201727/10/2017

Bibliographical note

For ECEL2017 http://www.academic-conferences.org/conferences/ecel/

 

Quantifying Feedback – Insights Into Peer Assessment Data

År: 2017

Quantifying Feedback – Insights Into Peer Assessment Data

Wind, D. K. & Jensen, U. A. 2017 Proceedings of the 12th International Conference on e-Learning. Academic Conferences and Publishing International, 10 p.

Publication: Research - peer-reviewArticle in proceedings – Annual report year: 2017

The act of producing content - for example in forms of written reports - is one of the most used methods for teaching and learning all the way from primary school to university. It is a learning tool which helps students relate their theories to practice. Getting relevant and helpful feedback on this work is important to ensure a good learning experience for the students. Providing this feedback is often a time-consuming job for the teacher. An effective way to learn is to teach others, and similarly give feedback on work done by others. One way to approach a combined solution to the above challenges, is to use peer assessment in the classroom which as a learning method has become more and more popular. In this paper we look at data collected using the web-based peer assessment system Peergrade. The dataset consists of over 350 courses at more than 20 educational institutions and with a total of more than 10,000 students. The students have together made more than 100,000 peer-evaluations of work by other students, and these evaluations together contain more than 10,000,000 words of text feedback. A key problem when using peer assessment is to ensure high quality feedback between peers. Feedback here can be a combination of quantitative / summative feedback (numerical) and qualitative / formative feedback (text). A lot of work has been done on validating and ensuring quality of quantitative feedback. We propose a way to let students evaluate the quality of the feedback they receive to obtain a quality measure for the feedback. We investigate this measure of feedback quality, which biases are present and what trends can be observed across the dataset. Using our measure of feedback quality, we investigate how it relates to various factors like the length of the feedback text, the number of spelling mistakes, how positive it is and measures of the student’s report-writing skills.
Original languageEnglish
Title of host publicationProceedings of the 12th International Conference on e-Learning
Number of pages10
PublisherAcademic Conferences and Publishing International
Publication date2017
ISBN (print)978-1-911218-35-7
StatePublished - 2017
Event12th International Conference on e-Learning - Orlando, United States

Conference

Conference12th International Conference on e-Learning
CountryUnited States
CityOrlando
Period01/06/201702/06/2017

 

DeepLoc: prediction of protein subcellular localization using deep learning

År: 2017

DeepLoc: prediction of protein subcellular localization using deep learning

Almagro Armenteros, J. J., Sønderby, C. K., Sønderby, S. K., Nielsen, H. & Winther, O. 2017 In : Bioinformatics. 33, 21, p. 3387-3395

Publication: Research - peer-reviewJournal article – Annual report year: 2017

The prediction of eukaryotic protein subcellular localization is a well-studied topic in bioinformatics due to its relevance in proteomics research. Many machine learning methods have been successfully applied in this task, but in most of them, predictions rely on annotation of homologues from knowledge databases. For novel proteins where no annotated homologues exist, and for predicting the effects of sequence variants, it is desirable to have methods for predicting protein properties from sequence information only. Here, we present a prediction algorithm using deep neural networks to predict protein subcellular localization relying only on sequence information. At its core, the prediction model uses a recurrent neural network that processes the entire protein sequence and an attention mechanism identifying protein regions important for the subcellular localization. The model was trained and tested on a protein dataset extracted from one of the latest UniProt releases, in which experimentally annotated proteins follow more stringent criteria than previously. We demonstrate that our model achieves a good accuracy (78% for 10 categories; 92% for membrane-bound or soluble), outperforming current state-of-the-art algorithms, including those relying on homology information. The method is available as a web server at http://www.cbs.dtu.dk/services/DeepLoc . Example code is available at https://github.com/JJAlmagro/subcellular_localization . The dataset is available at http://www.cbs.dtu.dk/services/DeepLoc/data.php . jjalma@dtu.dk.
Original languageEnglish
JournalBioinformatics
Volume33
Issue number21
Pages (from-to)3387-3395
ISSN1367-4803
DOIs
StatePublished - 2017

 

An introduction to Deep learning on biological sequence data - Examples and solutions

År: 2017

An introduction to Deep learning on biological sequence data - Examples and solutions

Jurtz, V. I., Johansen, A. R., Nielsen, M., Almagro Armenteros, J. J., Nielsen, H., Kaae Sønderby, C., Winther, O. & Kaae Sønderby, S. 2017 In : Bioinformatics. 33, 22, p. 3685-3690

Publication: Research - peer-reviewJournal article – Annual report year: 2017

Deep neural network architectures such as convolutional and long short-term memory networks have become increasingly popular as machine learning tools during the recent years. The availability of greater computational resources, more data, new algorithms for training deep models and easy to use libraries for implementation and training of neural networks are the drivers of this development. The use of deep learning has been especially successful in image recognition; and the development of tools, applications and code examples are in most cases centered within this field rather than within biology. Here, we aim to further the development of deep learning methods within biology by providing application examples and ready to apply and adapt code templates. Given such examples, we illustrate how architectures consisting of convolutional and long short-term memory neural networks can relatively easily be designed and trained to state-of-the-art performance on three biological sequence problems: prediction of subcellular localization, protein secondary structure and the binding of peptides to MHC Class II molecules. All implementations and datasets are available online to the scientific community at https://github.com/vanessajurtz/lasagne4bio . Supplementary data are available at Bioinformatics online.
Original languageEnglish
JournalBioinformatics
Volume33
Issue number22
Pages (from-to)3685-3690
ISSN1367-4803
DOIs
StatePublished - 2017

  PDF

Gradient distortions in EEG provide motion tracking during simultaneous EEG-fMRI

År: 2017

Gradient distortions in EEG provide motion tracking during simultaneous EEG-fMRI

Laustsen, M., Andersen, M., Madsen, K. H. & Hanson, L. G. 2017

Publication: Research - peer-reviewConference abstract for conference – Annual report year: 2017

Conference abstract, selected for oral presentation by Malte Laustsen.
Original languageEnglish
Publication date2017
Number of pages2
StatePublished - 2017
EventISMRM Workshop on Motion Correction in MRI & MRS - Cape Town, South Africa

Conference

ConferenceISMRM Workshop on Motion Correction in MRI & MRS
CountrySouth Africa
CityCape Town
Period08/09/201711/09/2017
Internet address

  PDF

Contact activity and dynamics of the social core

År: 2017

Contact activity and dynamics of the social core

Mones, E., Stopczynski, A. & Jørgensen, S. L. 2017 In : E P J Data Science. 6, 1, 16 p.

Publication: Research - peer-reviewJournal article – Annual report year: 2017

Humans interact through numerous communication channels to build and maintain social connections: they meet face-to-face, make phone calls or send text messages, and interact via social media. Although it is known that the network of physical contacts, for example, is distinct from the network arising from communication events via phone calls and instant messages, the extent to which these networks differ is not clear. We show here that the network structure of these channels show large structural variations. The various channels account for diverse relationships between pairs of individuals and the corresponding interaction patterns across channels differ to an extent that social ties cannot easily be reduced to a single layer. Each network of interactions, however, contains both central and peripheral individuals: central members are characterized by higher connectivity and can reach a large fraction of the network within a low number of steps, in contrast to the nodes on the periphery. The origin and purpose of each communication network also determine the role of their respective central members: highly connected individuals in the person-to-person networks interact with their environment in a regular manner, while members central in the social communication networks display irregular behavior with respect to their physical contacts and are more active through irregular social events. Our results suggest that due to the inherently different functions of communication channels, each one favors different social behaviors and different strategies for interacting with the environment. These findings can facilitate the understanding of the varying roles and impact individuals have on the population, which can further shed light on the prediction and prevention of epidemic outbreaks, or information propagation.
Original languageEnglish
JournalE P J Data Science
Volume6
Issue number1
Number of pages16
ISSN2193-1127
DOIs
StatePublished - 2017

  PDF

On the Keyhole Hypothesis: High Mutual Information between Ear and Scalp EEG

År: 2017

On the Keyhole Hypothesis: High Mutual Information between Ear and Scalp EEG

Mikkelsen, K. B., Kidmose, P. & Hansen, L. K. 2017 In : Frontiers in Human Neuroscience. 11, 9 p., 341

Publication: Research - peer-reviewJournal article – Annual report year: 2017

We propose and test the keyhole hypothesis that measurements from low dimensional EEG, such as ear-EEG reflect a broadly distributed set of neural processes. We formulate the keyhole hypothesis in information theoretical terms. The experimental investigation is based on legacy data consisting of 10 subjects exposed to a battery of stimuli, including alpha-attenuation, auditory onset, and mismatch-negativity responses and a new medium-long EEG experiment involving data acquisition during 13 h. Linear models were estimated to lower bound the scalp-to-ear capacity, i.e., predicting ear-EEG data from simultaneously recorded scalp EEG. A cross-validation procedure was employed to ensure unbiased estimates. We present several pieces of evidence in support of the keyhole hypothesis: There is a high mutual information between data acquired at scalp electrodes and through the ear-EEG "keyhole," furthermore we show that the view represented as a linear mapping is stable across both time and mental states. Specifically, we find that ear-EEG data can be predicted reliably from scalp EEG. We also address the reverse view, and demonstrate that large portions of the scalp EEG can be predicted from ear-EEG, with the highest predictability achieved in the temporal regions and when using ear-EEG electrodes with a common reference electrode.
Original languageEnglish
Article number341
JournalFrontiers in Human Neuroscience
Volume11
Number of pages9
ISSN1662-5161
DOIs
StatePublished - 2017

 

Active vibration-based structural health monitoring system for wind turbine blade: Demonstration on an operating Vestas V27 wind turbine

År: 2017

Active vibration-based structural health monitoring system for wind turbine blade: Demonstration on an operating Vestas V27 wind turbine

Tcherniak, D. & Mølgaard, L. L. 2017 In : Structural Health Monitoring. 16, 5, p. 536-550

Publication: Research - peer-reviewJournal article – Annual report year: 2017

This study presents a structural health monitoring system that is able to detect structural defects of wind turbine blade such as cracks, leading/trailing-edge opening, or delamination. It is shown that even small defects of at least 15 cm size can be detected remotely without stopping the wind turbine. The structural health monitoring system presented is vibration-based: mechanical energy is artificially introduced by means of an electromechanical actuator, whose plunger periodically hits the blade. The induced vibrations propagate along the blade and are picked up by accelerometers mounted along the blade. The vibrations in mid-range frequencies are utilized: this range is above the frequencies excited by blade–wind interaction, ensuring a good signal-to-noise ratio. At the same time, the corresponding wavelength is short enough to deliver required damage detection resolution and long enough to be able to propagate the entire blade length. This article demonstrates the system on a Vestas V27 wind turbine. One blade of the wind turbine was equipped with the system, and a 3.5-month monitoring campaign was conducted while the turbine was operating normally. During the campaign, a defect—a trailing-edge opening—was artificially introduced into the blade and its size was gradually increased from the original 15 to 45 cm. Using a semi-supervised learning algorithm, the system was able to detect even the smallest amount of damage while the wind turbine was operating under different weather conditions. This article provides detailed information about the instrumentation and the measurement campaign and explains the damage detection algorithm.
Original languageEnglish
JournalStructural Health Monitoring
Volume16
Issue number5
Pages (from-to)536-550
ISSN1475-9217
DOIs
StatePublished - 2017

  PDF

Infinite von Mises-Fisher Mixture Modeling of Whole Brain fMRI Data

År: 2017

Infinite von Mises-Fisher Mixture Modeling of Whole Brain fMRI Data

Røge, R., Madsen, K. H., Schmidt, M. N. & Mørup, M. 2017 In : Neural Computation. 29, 10, p. 2712-2741

Publication: Research - peer-reviewJournal article – Annual report year: 2017

Cluster analysis of functional magnetic resonance imaging (fMRI) data is often performed using gaussian mixture models, but when the time series are standardized such that the data reside on a hypersphere, this modeling assumption is questionable. The consequences of ignoring the underlying spherical manifold are rarely analyzed, in part due to the computational challenges imposed by directional statistics. In this letter, we discuss a Bayesian von Mises-Fisher (vMF) mixture model for data on the unit hypersphere and present an efficient inference procedure based on collapsed Markov chain Monte Carlo sampling. Comparing the vMF and gaussian mixture models on synthetic data, we demonstrate that the vMF model has a slight advantage inferring the true underlying clustering when compared to gaussian-based models on data generated from both a mixture of vMFs and a mixture of gaussians subsequently normalized. Thus, when performing model selection, the two models are not in agreement. Analyzing multisubject whole brain resting-state fMRI data from healthy adult subjects, we find that the vMF mixture model is considerably more reliable than the gaussian mixture model when comparing solutions across models trained on different groups of subjects, and again we find that the two models disagree on the optimal number of components. The analysis indicates that the fMRI data support more than a thousand clusters, and we confirm this is not a result of overfitting by demonstrating better prediction on data from held-out subjects. Our results highlight the utility of using directional statistics to model standardized fMRI data and demonstrate that whole brain segmentation of fMRI data requires a very large number of functional units in order to adequately account for the discernible statistical patterns in the data.
Original languageEnglish
JournalNeural Computation
Volume29
Issue number10
Pages (from-to)2712-2741
ISSN0899-7667
DOIs
StatePublished - 2017

 

Whole-brain functional connectivity predicted by indirect structural connections

År: 2017

Whole-brain functional connectivity predicted by indirect structural connections

Røge, R., Ambrosen, K. M. S., Albers, K. J., Eriksen, C. T., Liptrot, M. G., Schmidt, M. N., Madsen, K. H. & Mørup, M. 2017 Proceedings of 2017 International Workshop on Pattern Recognition in Neuroimaging . IEEE, p. 4 pp. (2017 International Workshop on Pattern Recognition in Neuroimaging (prni)).

Publication: Research - peer-reviewArticle in proceedings – Annual report year: 2017

Modern functional and diffusion magnetic resonance imaging (fMRI and dMRI) provide data from which macro-scale networks of functional and structural whole brain connectivity can be estimated. Although networks derived from these two modalities describe different properties of the human brain, they emerge from the same underlying brain organization, and functional communication is presumably mediated by structural connections. In this paper, we assess the structure-function relationship by evaluating how well functional connectivity can be predicted from structural graphs. Using high-resolution whole brain networks generated with varying density, we contrast the performance of several non-parametric link predictors that measure structural communication flow. While functional connectivity is not well predicted directly by structural connections, we show that superior predictions can be achieved by taking indirect structural pathways into account. In particular, we find that the length of the shortest structural path between brain regions is a good predictor of functional connectivity in sparse networks (density less than one percent), and that this improvement comes from integrating indirect pathways comprising up to three steps. Our results support the existence of important indirect relationships between structure and function, extending beyond the immediate direct structural connections that are typically investigated.
Original languageEnglish
Title of host publicationProceedings of 2017 International Workshop on Pattern Recognition in Neuroimaging
PublisherIEEE
Publication date2017
Pages4 pp.
DOIs
StatePublished - 2017
Event2017 International Workshop on Pattern Recognition in Neuroimaging - Toronto, Canada

Conference

Conference2017 International Workshop on Pattern Recognition in Neuroimaging
LocationUniversity of Toronto
CountryCanada
CityToronto
Period21/06/201723/06/2017
Series2017 International Workshop on Pattern Recognition in Neuroimaging (prni)

 

Second-Order Assortative Mixing in Social Networks

År: 2017

Second-Order Assortative Mixing in Social Networks

Zhou, S., Cox, I. & Hansen, L. K. 2017 Complex Networks Viii : Proceedings of the 8th Conference on Complex Networks Complenet 2017. Springer, p. 3-15 (Springer Proceedings in Complexity).

Publication: Research - peer-reviewArticle in proceedings – Annual report year: 2017

In a social network, the number of links of a node, or node degree, is often assumed as a proxy for the node’s importance or prominence within the network. It is known that social networks exhibit the (first-order) assortative mixing, i.e. if two nodes are connected, they tend to have similar node degrees, suggesting that people tend to mix with those of comparable prominence. In this paper, we report the second-order assortative mixing in social networks. If two nodes are connected, we measure the degree correlation between their most prominent neighbours, rather than between the two nodes themselves. We observe very strong second-order assortative mixing in social networks, often significantly stronger than the first-order assortative mixing. This suggests that if two people interact in a social network, then the importance of the most prominent person each knows is very likely to be the same. This is also true if we measure the average prominence of neighbours of the two people. This property is weaker or negative in non-social networks. We investigate a number of possible explanations for this property. However, none of them was found to provide an adequate explanation. We therefore conclude that second-order assortative mixing is a new property of social networks.
Original languageEnglish
Title of host publicationComplex Networks Viii : Proceedings of the 8th Conference on Complex Networks Complenet 2017
PublisherSpringer
Publication date2017
Pages3-15
ISBN (print)9783319542416
DOIs
StatePublished - 2017
Event8th Conference on Complex Networks Complenet 2017 - Dubrovnik, Croatia

Conference

Conference8th Conference on Complex Networks Complenet 2017
LocationInter University Center Dubrovnik
CountryCroatia
CityDubrovnik
Period21/03/201724/03/2017
SeriesSpringer Proceedings in Complexity
ISSN2213-8684

  PDF

Modeling dynamic functional connectivity using a wishart mixture model

År: 2017

Modeling dynamic functional connectivity using a wishart mixture model

Nielsen, S. F. V., Madsen, K. H., Schmidt, M. N. & Mørup, M. 2017 Proceedings of the 2017 International Workshop on Pattern Recognition in Neuroimaging. IEEE, p. 1-4 4 p. (2017 International Workshop on Pattern Recognition in Neuroimaging (prni)).

Publication: Research - peer-reviewArticle in proceedings – Annual report year: 2017

Dynamic functional connectivity (dFC) has recently become a popular way of tracking the temporal evolution of the brains functional integration. However, there does not seem to be a consensus on how to choose the complexity, i.e. number of brain states, and the time-scale of the dynamics, i.e. the window length. In this work we use the Wishart Mixture Model (WMM) as a probabilistic model for dFC based on variational inference. The framework admits arbitrary window lengths and number of dynamic components and includes the static one-component model as a special case. We exploit that the WMM framework provides model selection by quantifying models generalization to new data. We use this to quantify the number of states within a prespecified window length. We further propose a heuristic procedure for choosing the window length based on contrasting for each window length the predictive performance of dFC models to their static counterparts and choosing the window length having largest difference as most favorable for characterizing dFC. On synthetic data we find that generalizability is influenced by window length and signal-tonoise ratio. Too long windows cause dynamic states to be mixed together whereas short windows are more unstable and influenced by noise and we find that our heuristic correctly identifies an adequate level of complexity. On single subject resting state fMRI data we find that dynamic models generally outperform static models and using the proposed heuristic points to a windowlength of around 30 seconds provides largest difference between the predictive likelihood of static and dynamic FC.
Original languageEnglish
Title of host publicationProceedings of the 2017 International Workshop on Pattern Recognition in Neuroimaging
Number of pages4
PublisherIEEE
Publication date2017
Pages1-4
ISBN (print)978-1-5386-3159-1
DOIs
StatePublished - 2017
Event2017 International Workshop on Pattern Recognition in Neuroimaging - Toronto, Canada

Conference

Conference2017 International Workshop on Pattern Recognition in Neuroimaging
LocationUniversity of Toronto
CountryCanada
CityToronto
Period21/06/201723/06/2017
Series2017 International Workshop on Pattern Recognition in Neuroimaging (prni)