|
Activity Number:
|
285
- Statistical Inference for Probabilistic Graphical Models with Applications
|
|
Type:
|
Invited
|
|
Date/Time:
|
Wednesday, August 5, 2020 : 10:00 AM to 11:50 AM
|
|
Sponsor:
|
ENAR
|
|
Abstract #309258
|
|
|
Title:
|
Bayesian Structure Learning for Dynamic Brain Connectivity
|
|
Author(s):
|
Sanmi Koyejo*
|
|
Companies:
|
Department of Computer Science, University of Illinois at Urbana-Champaign
|
|
Keywords:
|
fMRI;
Brain Connectivity;
Structure Estimation
|
|
Abstract:
|
This talk will outline a Bayesian model for dynamic brain connectivity. Motivated by neuroscience priors, the model estimates covariances which vary smoothly over time, with an instantaneous decomposition into a collection of spatially sparse components – resulting in parsimonious and highly interpretable estimates of dynamic brain connectivity. Simulated results are presented to illustrate the performance of the model even when it is misspecified. For real brain imaging data with unknown ground truth, in addition to qualitative evaluation, we devise a simple classification task which suggests that the estimated brain networks better capture the underlying structure. The performance is further improved using a multimodal approach that combines fMRI with other brain imaging modalities.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2020 program
|
