Classification of Alzheimer's Dementia vs. Healthy subjects by studying structural disparities in fMRI Time-Series of DMN

TL;DR

This study introduces deviation from stochasticity ($DS$) as a biomarker for distinguishing Alzheimer's disease (AD) from healthy aging using resting-state fMRI time-series from the default mode network (DMN). An autoencoder-based, time-invariant representation is applied to multi-scale fMRI windows, with $DS$ computed via KL divergence-based dissimilarity and coefficient-of-variation metrics across ROIs (34 DMN ROIs). On the ADNI dataset (50 healthy controls, 50 AD), DS-based features yield high discrimination, achieving up to 95% accuracy with Random Forest/Gradient Boosting classifiers and ROC-AUC values up to 0.99, with SHAP analyses highlighting key ROIs such as the precuneus and prefrontal regions. The work suggests that DS captures disease-related stochasticity changes in DMN dynamics and offers a potential, non-invasive biomarker for early AD detection and monitoring, warranting further multi-site validation.

Abstract

Time series from different regions of interest (ROI) of default mode network (DMN) from Functional Magnetic Resonance Imaging (fMRI) can reveal significant differences between healthy and unhealthy people. Here, we propose the utility of an existing metric quantifying the lack/presence of structure in a signal called, "deviation from stochasticity" (DS) measure to characterize resting-state fMRI time series. The hypothesis is that differences in the level of structure in the time series can lead to discrimination between the subject groups. In this work, an autoencoder-based model is utilized to learn efficient representations of data by training the network to reconstruct its input data. The proposed methodology is applied on fMRI time series of 50 healthy individuals and 50 subjects with Alzheimer's Disease (AD), obtained from publicly available ADNI database. DS measure for healthy fMRI as expected turns out to be different compared to that of AD. Peak classification accuracy of 95% was obtained using Gradient Boosting classifier, using the DS measure applied on 100 subjects.

Figures (6)

• Figure 1: Proposed Methodology
• Figure 2: Scatter Plot separating AD and HC Subjects using CV1 and CV2 Values show discrimination between the 2 groups
• Figure 3: t-SNE Plot for Subjects based on DS Values clearly illustrates that 2 distinct clusters are formed, showing the goodness of the feature space
• Figure 4: SHAP based on DS Values
• Figure 5: DS Measure Difference between vmPFC 7 & 1 (right and left sides of brain) of AD Subject