PGAD: Prototype-Guided Adaptive Distillation for Multi-Modal Learning in AD Diagnosis
This addresses a critical issue for real-world clinical diagnosis of Alzheimer's Disease, where incomplete data is common, but the approach is incremental as it builds on existing multi-modal learning methods.
The paper tackles the problem of missing modalities in Alzheimer's Disease diagnosis by proposing a framework that incorporates incomplete multi-modal data into training, demonstrating significant performance improvements over state-of-the-art methods on the ADNI dataset with varying missing rates.
Missing modalities pose a major issue in Alzheimer's Disease (AD) diagnosis, as many subjects lack full imaging data due to cost and clinical constraints. While multi-modal learning leverages complementary information, most existing methods train only on complete data, ignoring the large proportion of incomplete samples in real-world datasets like ADNI. This reduces the effective training set and limits the full use of valuable medical data. While some methods incorporate incomplete samples, they fail to effectively address inter-modal feature alignment and knowledge transfer challenges under high missing rates. To address this, we propose a Prototype-Guided Adaptive Distillation (PGAD) framework that directly incorporates incomplete multi-modal data into training. PGAD enhances missing modality representations through prototype matching and balances learning with a dynamic sampling strategy. We validate PGAD on the ADNI dataset with varying missing rates (20%, 50%, and 70%) and demonstrate that it significantly outperforms state-of-the-art approaches. Ablation studies confirm the effectiveness of prototype matching and adaptive sampling, highlighting the potential of our framework for robust and scalable AD diagnosis in real-world clinical settings.