Explaining deep learning-based representations of resting state functional connectivity data

Focusing on interpreting nonlinear patterns in autism spectrum disorder

Datos Bibliográficos

ID15519179
AutoresYoung-Geun Kim (0000-0001-8910-1227, Columbia University Irving Medical Center), Orren Ravid (Columbia University Irving Medical Center), Xinyuan Zheng (0009-0003-2808-5907, New York State Office of Mental Health), Yoojean Kim (Columbia University Irving Medical Center), Yuval Neria (0000-0001-5223-4333, Columbia University Irving Medical Center), Seonjoo Lee (0000-0003-3177-6357, New York State Office of Mental Health), Xiaofu He (0000-0002-7401-1656, Columbia University Irving Medical Center, autor de correspondencia), Xi Zhu (0000-0003-1342-2221, Columbia University Irving Medical Center, autor de correspondencia)
Año2024
Volumen15
Páginas1397093-1397093
Fecha de publicación2024-05-20
Peer ReviewedSí
Open AccessSí
TipoARTICLE
RevistaFrontiers in Psychiatry (JOURNAL)
Identificadores de la revistaISSN: 1664-0640 • E-ISSN: 1664-0640
EditorialFrontiers Media (PUBLISHER • CH)
DOI10.3389/fpsyt.2024.1397093
PMID38832332
OpenAlexW4398138592
IdiomaEN
Referencias citadas23

This study introduced latent contribution scores to interpret nonlinear patterns identified by VAEs. These scores effectively capture changes in each observed rsFC feature as the estimated latent representation changes, enabling an explainable deep learning model that better understands the underlying neural mechanisms of ASD

Autism · Autism spectrum disorder · Cognitive psychology · Developmental psychology · Functional connectivity · Machine learning · Pattern recognition (psychology · Resting state fMRI · Computer Science · EEG and Brain-Computer Interfaces · Functional Brain Connectivity Studies · Neural dynamics and brain function · Neuroscience · Psychology · Artificial Intelligence