Integrative Variational Autoencoders for Generative Modeling of an Image Outcome with Multiple Input Images
Year: 2024
Type: preprint
Abstract: Understanding relationships across multiple imaging modalities is central to neuroimaging research. We introduce the Integrative Variational Autoencoder (InVA), the first hierarchical VAE framework for image-on-image regression in multimodal neuroimaging. Unlike standard VAEs, which are not designed for predictive integration across modalities, InVA models outcome images as functions of both shared and modality-specific features. This flexible, data-driven approach avoids rigid assumptions of classical tensor regression and outperforms conventional VAEs and nonlinear models such as BART. As a key application, InVA accurately predicts costly PET scans from structural MRI, offering an efficient and powerful tool for multimodal neuroimaging.
Source: arXiv (Cornell University)
Authors: Lei, Bowen, Jeon, Yeseul, Guhaniyogi, Rajarshi, Scheffler, Aaron, Mallick, Bani, Initiatives, Alzheimer's Disease Neuroimaging
Language: English
Subfield: Cognitive Neuroscience
Field: Neuroscience
Domain: Life Sciences
Open Access status: green
Funders: National Science Foundation, U.S. Department of Defense, Alzheimer's Association, Bristol-Myers Squibb, Eli Lilly and Company, Pfizer, Biogen, University of Southern California, Meso Scale Diagnostics, Alzheimer's Disease Neuroimaging Initiative, BioClinica, Novartis Pharmaceuticals Corporation, Northern California Institute for Research and Education, Eisai, Servier, H. Lundbeck A/S, National Institutes of Health, Genentech, Canadian Institutes of Health Research
Awards: G1298807586, G1554142629, G1593117279, G1650819308, G1880935189, Alzheimer's Disease Neuroimaging Initiative, G2100670906, G2152643862, G2189010064, G2340207221, G2423387114, G2471646051, G2522063444, Collaborative Research: Use of Random Compression Matrices For Scalable Inference in High Dimensional Structured Regressions, G2835601961, G29951154, G3131374986, G3216574877, Collaborative Research: Use of Random Compression Matrices For Scalable Inference in High Dimensional Structured Regressions, G3553156159, G3710456697, G3752451908, G3870774570, G3908963977, G4119534169, G43180419, G4475667458, G4800177743, G5207404921, G5209990037, G528601746, G5365730499, EFFECTS OF TRAUMATIC BRAIN INJURY (TBI) AND POST TRAUMATIC STRESS DISORDER (PTSD) ON ALZHEIMER'S DISEASE (AD) IN VETERANS USING IMAGING AND BIOMARKERS IN THE AD NEUROIMAGING INITIATIVE (ADNI)", G5654639750, G5740349702, G5850682332, G613619302, G6173816828, G6210868510, G6371410416, G6412110892, G6447063303, G6792506723, G6899163935, G72415811, Bayesian Object-Oriented Modeling of Multi-Modal Imaging Data, G764699806, G7799614282, G781725486, G8228018387, G8281378126, G8301434335, G839208290, G8852128998, G8895907669, G8897063270, G8957659042, G9000039018, G9000122335
OpenAlex ID: W4391591318