Optimally mapping multivariate regional brain age and clinical severity patterns in Alzheimer's disease using AgeNet‐SHAP explainable AI approach

The explainable AI method, AgeNet-SHAP, demonstrated good performance in brain age prediction based on a large MRI dataset, along with identifying the hierarchy of multivariate brain regional associations with this prediction. Additionally, this method captured the brain regional associations with AD severity. These results collectively facilitate data-driven predictive modelling approaches for disease progression, diagnostics, prognostics, and personalized medicine efforts.