Sidsel Borup Winther: Inferring properties of neural microstructures to improve our understanding of the brain
Our brain function sculpts who we are and how we perceive the world around us. Hence, when disorders like multiple sclerosis, Alzheimer’s, schizophrenia, and traumatic brain injury are causing changes in the brain function it is crucial for the lives of the affected. Learning the underlying mechanisms of the changes, is key to understanding how exactly the brain is affected and how we can obtain correct and early diagnosis.
To study and discover these mechanisms we must be able to acquire noninvasive measurements of microstructural properties in living brains. The most promising method to do so, is diffusion weighted magnetic resonance imaging (DWI). By measuring the diffusion of atomic nuclei within the brain, the signal is sensitive to shapes at the microstructural scale. But due to limited spatial resolution each voxel will contain different cell types with varying sizes and orientations, and inverse modeling of the signal is therefore necessary in order to make inference of parameters describing the underlying microstructural properties. Modeling of the signal acquired for the highly complex biological structures is however very challenging, and a fully adequate method is yet to be developed.
The aim of this project is to push the frontiers of microstructure modeling by incorporating virtual simulations, multi modal integration, biophysical modeling, and machine learning modelling.
The project is carried out in cooperation with Danish Research Center For Magnetic Resonance (DRCMR) and École Polytechnique Fédérale de Lausanne (EPFL).
Figure: Alexander, D. C., Dyrby, T. B., Nilsson, M., & Zhang, H. (2018). Imaging brain microstructure with diffusion MRI: Practicality and applications: practicality and applications. N M R in Biomedicine, 32(4), [e3841]. https://doi.org/10.1002/nbm.3841