A Demographic-Conditioned Variational Autoencoder for fMRI Distribution Sampling and Removal of Confounds.

Anton OrlichenkoGang QuZiyu ZhouAnqi LiuHong-Wen DengZhengming DingJulia M StephenTony W Wilson Vince D CalhounYu-Ping Wang

Published in: bioRxiv : the preprint server for biology (2024)

Our DemoVAE model allows for generation of high quality synthetic data conditioned on subject demographics as well as the removal of the confounding effects of demographics. We identify that FC-based prediction tasks are highly influenced by demographic confounds.

Keyphrases

resting state
functional connectivity
electronic health record
working memory
data analysis
deep learning