Geometric Approaches for Processing Brain Connectomes
Автор: WiDS Worldwide
Загружено: 2025-01-09
Просмотров: 884
We demonstrate how to apply geometric machine learning approaches to brain connectome data and how this can be used to classify brains as having Schizophrenia or not. We use data from the MSLP 2014 Schizophrenia Challenge. The dataset corresponds to the Functional Connectivity Networks (FCN) extracted from resting-state fMRIs of 86 patients at 28 regions of interest. Patients are separated in two classes: schizophrenic and control. Roughly, an FCN corresponds to a correlation matrix and can be seen as a point on the manifold of Symmetric Positive-Definite (SPD) matrices. Thus, tools for machine learning on manifolds are useful for comparing, classifying, and characterizing FCN data.
By Adele Myers, Physics Ph.D. Candidate at UC Santa Barbara
Доступные форматы для скачивания:
Скачать видео mp4
-
Информация по загрузке:
