Multi-modal Imputation for Alzheimer's Disease Classification

2026-01-28

AI Summary

This study addresses the challenge of missing modalities in multimodal neuroimaging for Alzheimer’s disease classification, where clinical utility is often limited by incomplete data such as absent diffusion-weighted imaging (DWI) despite available T1-weighted MRI. To tackle this, the work proposes the first application of a conditional denoising diffusion probabilistic model to synthesize high-quality DWI from T1 scans, enabling robust classification across three disease states. Through systematic evaluation of both unimodal and bimodal classifiers under various imputation scenarios, the method demonstrates significant improvements in downstream classification performance, particularly enhancing sensitivity for underrepresented minority classes. These findings underscore the efficacy and potential of diffusion models in handling missing data through cross-modal synthesis in multimodal medical imaging.

Technology Category

• Machine Learning: Multimodal Learning
• Computer Vision: Multi-modal Vision
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Deep learning has been successful in predicting neurodegenerative disorders, such as Alzheimer's disease, from magnetic resonance imaging (MRI). Combining multiple imaging modalities, such as T1-weighted (T1) and diffusion-weighted imaging (DWI) scans, can increase diagnostic performance. However, complete multimodal datasets are not always available. We use a conditional denoising diffusion probabilistic model to impute missing DWI scans from T1 scans. We perform extensive experiments to evaluate whether such imputation improves the accuracy of uni-modal and bi-modal deep learning models for 3-way Alzheimer's disease classification-cognitively normal, mild cognitive impairment, and Alzheimer's disease. We observe improvements in several metrics, particularly those sensitive to minority classes, for several imputation configurations.

Problem

• Alzheimer's disease
• multi-modal imputation
• missing data
• MRI
• classification

Innovation

• conditional denoising diffusion
• multi-modal imputation
• Alzheimer's disease classification
• MRI
• deep learning