Explainable Brain Age Prediction using coVariance Neural Networks

TL;DR

This paper introduces an explainable brain age prediction method using coVariance neural networks that enhances interpretability by identifying contributing brain regions and leveraging eigenvectors of anatomical covariance.

Contribution

The study presents a novel, anatomically interpretable brain age prediction framework using coVariance neural networks, addressing transparency issues in existing models.

Findings

VNNs can identify brain regions contributing to elevated brain age gap in AD.

Interpretability depends on exploiting specific eigenvectors of the anatomical covariance matrix.

The framework extends beyond simple brain age gap metrics in Alzheimer's disease.

Abstract

In computational neuroscience, there has been an increased interest in developing machine learning algorithms that leverage brain imaging data to provide estimates of "brain age" for an individual. Importantly, the discordance between brain age and chronological age (referred to as "brain age gap") can capture accelerated aging due to adverse health conditions and therefore, can reflect increased vulnerability towards neurological disease or cognitive impairments. However, widespread adoption of brain age for clinical decision support has been hindered due to lack of transparency and methodological justifications in most existing brain age prediction algorithms. In this paper, we leverage coVariance neural networks (VNN) to propose an explanation-driven and anatomically interpretable framework for brain age prediction using cortical thickness features. Specifically, our brain age prediction framework extends beyond the coarse metric of brain age gap in Alzheimer's disease (AD) and we make two important observations: (i) VNNs can assign anatomical interpretability to elevated brain age gap in AD by identifying contributing brain regions, (ii) the interpretability offered by VNNs is contingent on their ability to exploit specific eigenvectors of the anatomical covariance matrix. Together, these observations facilitate an explainable and anatomically interpretable perspective to the task of brain age prediction.