Explainable machine learning to predict and differentiate Alzheimer's progression by gender

Our primary objective is to develop an explainable Machine Learning (ML) model to support experts in predicting the conversion/onset of Alzheimer's disease (AD) and to explore differences between genders. Our intention to use only affordable neuropsychological tests. The findings confirm the gender-specific differences in AD progression and the relevance of specific cognitive measures in detecting individual's evolution in AD. In particular the delayed recall of logical and verbal memory as well as MMSE have been shown to be predictive of AD. Further research is needed to confirm and explore these differences, and to develop targeted interventions for individuals at risk for AD. This research demonstrates that explainable ML can effectively predict AD progression trajectories using not-expensive neuropsychological tests as features.