Progression models for imaging data with Longitudinal Variational Auto Encoders - INRIA - Institut National de Recherche en Informatique et en Automatique Access content directly
Conference Papers Year : 2022

Progression models for imaging data with Longitudinal Variational Auto Encoders

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.
Fichier principal
Vignette du fichier
MICCAI_2022___Accepted_version (2).pdf (3.82 Mo) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-03701632 , version 1 (22-06-2022)

Identifiers

  • HAL Id : hal-03701632 , version 1

Cite

Benoît Sauty, Stanley Durrleman​. Progression models for imaging data with Longitudinal Variational Auto Encoders. MICCAI 2022, International Conference on Medical Image Computing and Computer Assisted Intervention, Sep 2022, Singapore, Singapore. ⟨hal-03701632⟩
237 View
851 Download

Share

Gmail Mastodon Facebook X LinkedIn More