Graph Theory and GNNs to Unravel the Topographical Organization of Brain Lesions in Variants of Alzheimer's Disease Progression

arXiv:2403.00636v1 Announce Type: cross
Abstract: This study utilizes graph theory and deep learning to assess variations in Alzheimer's disease (AD) neuropathologies, focusing on classic (cAD) and rapid (rpAD) progression forms. It analyses the distribution of amyloid plaques and tau tangles in postmortem brain tissues. Histopathological images are converted into tau-pathology-based graphs, and derived metrics are used for statistical analysis and in machine learning classifiers. These classifiers incorporate SHAP value explainability to differentiate between cAD and rpAD. Graph neural networks (GNNs) …

abstract alzheimer's arxiv brain cad cs.cv deep learning disease distribution eess.iv forms gnns graph organization study theory type variants