A Neurosymbolic Approach for Explainable Early Diagnosis of Alzheimer's Disease

2026-07-31

AI Summary

This work proposes the first neurosymbolic approach to automatically construct interpretable clinical knowledge graphs from raw audio recordings of verbal fluency tests for Alzheimer’s disease biomarker identification. Addressing the inefficiency and limited scalability of traditional methods that rely on manual transcription and expert analysis, the system leverages pretrained foundation models to extract linguistic features and clinical variables, followed by qualitative relationship inference through a Bayesian network. The resulting framework not only accurately recapitulates established medical patterns but also uncovers novel associations among linguistic markers, offering an interpretable and scalable solution for early-stage Alzheimer’s screening.

Abstract

Identifying reliable Alzheimer's disease (AD) markers typically requires manual, labor-intensive transcription and expert analysis, limiting its scale. We introduce an automated pipeline that extracts qualitative knowledge about potential AD progression indicators directly from audio recordings of verbal fluency tests. Our method uses pretrained foundation models to process raw audio and extract clinically relevant variables to construct a Bayesian Network (BN); this BN is used to reason about the AD progression markers and infer their qualitative relationships. Our system successfully recovers known clinical knowledge and identifies novel relationships between linguistic markers.

Problem

• Alzheimer's disease
• early diagnosis
• biomarkers
• verbal fluency tests
• explainability

Innovation

• neurosymbolic
• Bayesian Network
• Alzheimer's disease
• explainable AI
• verbal fluency test