Exploring Nutritional Impact on Alzheimer's Mortality: An Explainable AI Approach

TL;DR

This study investigates how nutritional status relates to Alzheimer's disease mortality by leveraging NHANES III data and an explainable AI framework. It compares multiple ML models and uses SHAP to identify key nutritional biomarkers, selecting a Random Forest as the base model due to its predictive strength and interpretability. Vitamin B12 and 3-Methylhistidine emerge as consistent top predictors across comparisons, with SHAP providing mechanistic insights into how nutrition influences AD mortality versus Non-AD, heart disease, and cancer. While promising, the work acknowledges limitations such as sample size and missing-data handling, outlining plans for larger datasets and improved imputation to strengthen causal interpretation and translational relevance.

Abstract

This article uses machine learning (ML) and explainable artificial intelligence (XAI) techniques to investigate the relationship between nutritional status and mortality rates associated with Alzheimers disease (AD). The Third National Health and Nutrition Examination Survey (NHANES III) database is employed for analysis. The random forest model is selected as the base model for XAI analysis, and the Shapley Additive Explanations (SHAP) method is used to assess feature importance. The results highlight significant nutritional factors such as serum vitamin B12 and glycated hemoglobin. The study demonstrates the effectiveness of random forests in predicting AD mortality compared to other diseases. This research provides insights into the impact of nutrition on AD and contributes to a deeper understanding of disease progression.

Figures (1)

• Figure 1: Diagram of designed analysis approach (NC = Normal Control).