Self-Supervised Longitudinal Neighbourhood Embedding.

ABSTRACT: Longitudinal MRIs are often used to capture the gradual deterioration of brain structure and function caused by aging or neurological diseases. Analyzing this data via machine learning generally requires a large number of ground-truth labels, which are often missing or expensive to obtain. Reducing the need for labels, we propose a self-supervised strategy for representation learning named Longitudinal Neighborhood Embedding (LNE). Motivated by concepts in contrastive learning, LNE explicitly models the similarity between trajectory vectors across different subjects. We do so by building a graph in each training iteration defining neighborhoods in the latent space so that the progression direction of a subject follows the direction of its neighbors. This results in a smooth trajectory fiel

SUBMITTER: Ouyang J 

PROVIDER: S-EPMC9204645 | biostudies-literature | 2021 Sep-Oct

REPOSITORIES: biostudies-literature