Konrad Wagstyl : “BigBrain: Deep learning of cerebral cortex microstructure”

December 19, 2018
(We may stop the seminar at 13:00, but will continue the discussion with the speaker until 13:15.)
CiNet 1F Conference Room

“BigBrain: Deep learning of cerebral cortex microstructure”

Konrad Wagstyl
Postdoctoral Research Associate
Department of Psychiatry
University of Cambridge

Host : Hiromasa Takemura (Kida Group)

The cerebral cortex has a layered microstructure that varies depending on cortical area, local geometry and is altered by neurological and psychiatric pathologies. In this talk I will discuss the application of deep learning to analysing ex vivo and in vivo cortical microstructure. First I will explore how deep learning can be used to segment the layers and areas of the cerebral cortex of the BigBrain, a high resolution 3D histological dataset. I will then show how these can be used to interpret in vivo patterns of cortical structure measured using MRI. Finally, I will show how we are using deep learning to detect and segment microstructural lesions in hospital MRI scans of patients with epilepsy, caused by focal cortical dysplasias. This project has involved an open science approach and now has 23 epilepsy centres worldwide collaborating to create a cohort of patients.