An explainable framework for the relationship between dementia and glucose metabolism patterns

TL;DR

This paper introduces an explainable semi-supervised VAE framework that aligns neuroimaging features with dementia biomarkers, enabling interpretable analysis of Alzheimer's disease progression from PET scans.

Contribution

The proposed framework uniquely combines semi-supervised learning with flexible similarity regularization to interpret neuroimaging data in relation to dementia biomarkers.

Findings

Reduced metabolism in hippocampus and key networks correlates with cognitive impairment.

Latent variables effectively capture disease-related patterns and confounds.

The approach aligns neuroimaging features with established Alzheimer's biomarkers.

Abstract

High-dimensional neuroimaging data presents challenges for assessing neurodegenerative diseases due to complex non-linear relationships. Variational Autoencoders (VAEs) can encode scans into lower-dimensional latent spaces capturing disease-relevant features. We propose a semi-supervised VAE framework with a flexible similarity regularization term that aligns selected latent variables with clinical or biomarker measures of dementia progression. This allows adapting the similarity metric and supervised variables to specific goals or available data. We demonstrate the approach using PET scans from the Alzheimer's Disease Neuroimaging Initiative (ADNI), guiding the first latent dimension to align with a cognitive score. Using this supervised latent variable, we generate average reconstructions across levels of cognitive impairment. Voxel-wise GLM analysis reveals reduced metabolism in key regions, mainly the hippocampus, and within major Resting State Networks, particularly the Default Mode and Central Executive Networks. The remaining latent variables encode affine transformations and intensity variations, capturing confounds such as inter-subject variability and site effects. Our framework effectively extracts disease-related patterns aligned with established Alzheimer's biomarkers, offering an interpretable and adaptable tool for studying neurodegenerative progression.