AD-Reasoning: Multimodal Guideline-Guided Reasoning for Alzheimer's Disease Diagnosis

TL;DR

AD-Reasoning is a multimodal framework that integrates neuroimaging and clinical data with guideline-based reasoning to improve Alzheimer's diagnosis accuracy and transparency, using rule-based verification and reinforcement learning.

Contribution

The paper introduces AD-Reasoning, a novel multimodal reasoning framework that aligns diagnoses with clinical guidelines and enhances interpretability in Alzheimer's disease detection.

Findings

Achieves state-of-the-art diagnostic accuracy on AD-MultiSense dataset.

Produces structured, guideline-consistent rationales that improve transparency.

Demonstrates effective multimodal data fusion with reinforcement fine-tuning.

Abstract

Alzheimer's disease (AD) diagnosis requires integrating neuroimaging with heterogeneous clinical evidence and reasoning under established criteria, yet most multimodal models remain opaque and weakly guideline-aligned. We present AD-Reasoning, a multimodal framework that couples structural MRI with six clinical modalities and a rule-based verifier to generate structured, NIA-AA-consistent diagnoses. AD-Reasoning combines modality-specific encoders, bidirectional cross-attention fusion, and reinforcement fine-tuning with verifiable rewards that enforce output format, guideline evidence coverage, and reasoning--decision consistency. We also release AD-MultiSense, a 10,378-visit multimodal QA dataset with guideline-validated rationales built from ADNI/AIBL. On AD-MultiSense, AD-Reasoning achieves state-of-the-art diagnostic accuracy and produces structured rationales that improve transparency over recent baselines, while providing transparent rationales.