TY - GEN
T1 - Improving Normative Modeling for Multi-Modal Neuroimaging Data Using Mixture-of-Product-of-Experts Variational Autoencoders
AU - Kumar, Sayantan
AU - Payne, Philip
AU - Sotiras, Aristeidis
N1 - Publisher Copyright:
© 2024 IEEE.
PY - 2024
Y1 - 2024
N2 - Normative models in neuroimaging learn the brain patterns of healthy population distribution and estimate how disease subjects like Alzheimer's Disease (AD) deviate from the norm. Existing variational autoencoder (VAE)-based normative models using multimodal neuroimaging data aggregate information from multiple modalities by estimating product or averaging of unimodal latent posteriors. This can often lead to uninformative joint latent distributions which affects the estimation of subject-level deviations. In this work, we addressed the prior limitations by adopting the Mixture-of-Product-of-Experts (MoPoE) technique which allows better modelling of the joint latent posterior. Our model labelled subjects as outliers by calculating deviations from the multimodal latent space. Further, we identified which latent dimensions and brain regions were associated with abnormal deviations due to AD pathology.
AB - Normative models in neuroimaging learn the brain patterns of healthy population distribution and estimate how disease subjects like Alzheimer's Disease (AD) deviate from the norm. Existing variational autoencoder (VAE)-based normative models using multimodal neuroimaging data aggregate information from multiple modalities by estimating product or averaging of unimodal latent posteriors. This can often lead to uninformative joint latent distributions which affects the estimation of subject-level deviations. In this work, we addressed the prior limitations by adopting the Mixture-of-Product-of-Experts (MoPoE) technique which allows better modelling of the joint latent posterior. Our model labelled subjects as outliers by calculating deviations from the multimodal latent space. Further, we identified which latent dimensions and brain regions were associated with abnormal deviations due to AD pathology.
KW - mixture-of-product-of-experts
KW - multimodal
KW - normative modelling
KW - variational autoencoders
UR - http://www.scopus.com/inward/record.url?scp=85203332825&partnerID=8YFLogxK
U2 - 10.1109/ISBI56570.2024.10635897
DO - 10.1109/ISBI56570.2024.10635897
M3 - Conference contribution
AN - SCOPUS:85203332825
T3 - Proceedings - International Symposium on Biomedical Imaging
BT - IEEE International Symposium on Biomedical Imaging, ISBI 2024 - Conference Proceedings
PB - IEEE Computer Society
T2 - 21st IEEE International Symposium on Biomedical Imaging, ISBI 2024
Y2 - 27 May 2024 through 30 May 2024
ER -