Collaborative Attention and Consistent-Guided Fusion of MRI and PET for Alzheimer's Disease Diagnosis

TL;DR

This paper presents a novel multimodal fusion framework for Alzheimer's diagnosis using MRI and PET, emphasizing modality-specific features and distribution alignment to improve classification accuracy.

Contribution

The proposed model introduces a learnable parameter representation, shared and modality-specific encoders, and a consistency-guided mechanism for better multimodal fusion.

Findings

Achieves superior diagnostic accuracy on ADNI dataset

Effectively preserves modality-specific features

Aligns latent distributions across modalities

Abstract

Alzheimer's disease (AD) is the most prevalent form of dementia, and its early diagnosis is essential for slowing disease progression. Recent studies on multimodal neuroimaging fusion using MRI and PET have achieved promising results by integrating multi-scale complementary features. However, most existing approaches primarily emphasize cross-modal complementarity while overlooking the diagnostic importance of modality-specific features. In addition, the inherent distributional differences between modalities often lead to biased and noisy representations, degrading classification performance. To address these challenges, we propose a Collaborative Attention and Consistent-Guided Fusion framework for MRI and PET based AD diagnosis. The proposed model introduces a learnable parameter representation (LPR) block to compensate for missing modality information, followed by a shared encoder and modality-independent encoders to preserve both shared and specific representations. Furthermore, a consistency-guided mechanism is employed to explicitly align the latent distributions across modalities. Experimental results on the ADNI dataset demonstrate that our method achieves superior diagnostic performance compared with existing fusion strategies.