CONTRASTIVE LEARNING FOR REGRESSION IN MULTI-SITE BRAIN AGE PREDICTION - Laboratoire Traitement et Communication de l'Information Access content directly
Conference Papers Year : 2023

CONTRASTIVE LEARNING FOR REGRESSION IN MULTI-SITE BRAIN AGE PREDICTION

Abstract

Building accurate Deep Learning (DL) models for brain age prediction is a very relevant topic in neuroimaging, as it could help better understand neurodegenerative disorders and find new biomarkers. To estimate accurate and generalizable models, large datasets have been collected, which are often multi-site and multi-scanner. This large heterogeneity negatively affects the generalization performance of DL models since they are prone to overfit site-related noise. Recently, contrastive learning approaches have been shown to be more robust against noise in data or labels. For this reason, we propose a novel contrastive learning regression loss for robust brain age prediction using MRI scans. Our method achieves state-of-the-art performance on the OpenBHB challenge, yielding the best generalization capability and robustness to site-related noise.
Fichier principal
Vignette du fichier
ISBI_2023___brain_age_mri.pdf (373.37 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03993158 , version 1 (16-02-2023)

Identifiers

  • HAL Id : hal-03993158 , version 1

Cite

Carlo Alberto Barbano, Benoit Dufumier, Edouard Duchesnay, Marco Grangetto, Pietro Gori. CONTRASTIVE LEARNING FOR REGRESSION IN MULTI-SITE BRAIN AGE PREDICTION. IEEE ISBI (International Symposium on Biomedical Imaging ), Apr 2023, Cartagena de Indias, Colombia. ⟨hal-03993158⟩
73 View
150 Download

Share

Gmail Facebook X LinkedIn More