Progression models for imaging data with Longitudinal Variational Auto Encoders

By Benoît Sauty, Stanley Durrleman

Keywords: Disease Progression Modelling Medical imaging

This work presents a framework to predict how the brain will look in the future from imaging scans at a given time. This is very important as it allows us to predict how patients with neurodegenerative diseases will evolve before the disease really kicks in !

Published in International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) 2022 the 19/09/2022

Full title Progression models for imaging data with Longitudinal Variational Auto Encoders

Article abstract:

Disease progression models are crucial to understanding degenerative diseases. Mixed-effects models have been consistently used to model clinical assessments or biomarkers extracted from medical images, allowing missing data imputation and prediction at any timepoint. However, such progression models have seldom been used for entire medical images. In this work, a Variational Auto Encoder is coupled with a temporal linear mixed-effect model to learn a latent representation of the data such that individual trajectories follow straight lines over time and are characterised by a few interpretable parameters. A Monte Carlo estimator is devised to iteratively optimize the networks and the statistical model. We apply this method on a synthetic data set to illustrate the disentanglement between time dependant changes and inter-subjects variability, as well as the predictive capabilities of the method. We then apply it to 3D MRI and FDG-PET data from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) to recover well documented patterns of structural and metabolic alterations of the brain.