Longitudinal Self-Supervised Learning

TL;DR

This paper introduces Longitudinal Self-Supervised Learning (LSSL), a method that disentangles brain age from MRI data over time without labels, improving neuroimaging analysis and classification accuracy.

Contribution

LSSL is a novel self-supervised approach that explicitly disentangles longitudinal factors like brain age from MRI representations, enhancing analysis of neurodegenerative data.

Findings

LSSL effectively extracts brain-age information from longitudinal MRI data.

LSSL improves classification accuracy and convergence speed.

LSSL reveals neurodegenerative and neuropsychological disorder characteristics.

Abstract

Machine learning analysis of longitudinal neuroimaging data is typically based on supervised learning, which requires a large number of ground-truth labels to be informative. As ground-truth labels are often missing or expensive to obtain in neuroscience, we avoid them in our analysis by combing factor disentanglement with self-supervised learning to identify changes and consistencies across the multiple MRIs acquired of each individual over time. Specifically, we propose a new definition of disentanglement by formulating a multivariate mapping between factors (e.g., brain age) associated with an MRI and a latent image representation. Then, factors that evolve across acquisitions of longitudinal sequences are disentangled from that mapping by self-supervised learning in such a way that changes in a single factor induce change along one direction in the representation space. We implement this model, named Longitudinal Self-Supervised Learning (LSSL), via a standard autoencoding structure with a cosine loss to disentangle brain age from the image representation. We apply LSSL to two longitudinal neuroimaging studies to highlight its strength in extracting the brain-age information from MRI and revealing informative characteristics associated with neurodegenerative and neuropsychological disorders. Moreover, the representations learned by LSSL facilitate supervised classification by recording faster convergence and higher (or similar) prediction accuracy compared to several other representation learning techniques.