Multi-Modal Longitudinal Framework for Alzheimer's Disease Diagnosis Using DPTPSAN and SHAP-Based Interpretability
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
This paper is addressing AD diagnosis on pre-clinical stage by considering the longitudinal multimodal biomarker studies which reveal that the continuum of Alzheimer's disease (AD) includes a long latent phase, referred to as preclinical AD, which precedes the onset of symptoms by decades. Diagnosis treatment during the preclinical AD phase offers an optimal opportunity for slowing the progression of disease. Despite significant advancements in the field of Alzheimer's disease (AD) diagnosis, current diagnostic models still face several challenges that hinder their clinical applicability. Most existing methods focus either on neuroimaging or clinical data in isolation, without effectively leveraging the complementary strengths of multi-modal data. Moreover, many diagnostic frameworks rely on cross-sectional data, ignoring the temporal dynamics that are crucial for understanding the progression of AD. The integration of longitudinal clinical data with imaging information remains underexplored, and synthetic adjustment of such heterogeneous data is rarely addressed in the literature. Furthermore, issues like class imbalance, suboptimal feature fusion, limited interpretability, and lack of real-world deployment mechanisms-such as mobile-based tools-further exacerbate the challenge.
Fields of Study & Subjects
Dementia and Cognitive Impairment Research
Artificial Intelligence in Healthcare
Alzheimer's disease research and treatments
Publication details
- Journal / source
- International Conference on Emerging Trends and Innovations in ICT (ICEI)
- Pages
- 1–6
- Publication date
- 9 January 2026
- Language
- EN