Artificial Intelligence–Enhanced Multi-Algorithm R Shiny Application for Predictive Modeling and Analytics: Case Study of Alzheimer Disease Diagnostics

BACKGROUND: AI has demonstrated superior diagnostic accuracy compared to medical practitioners, highlighting its growing importance in healthcare. SMART-Pred (Shiny Multi-Algorithm R Tool for Predictive Modeling) is an innovative AI-based application for Alzheimer's disease (AD) prediction using handwriting analysis. OBJECTIVE: To develop and evaluate a non-invasive, cost-effective AI tool for early AD detection, addressing the need for accessible and accurate screening methods. METHODS: The study employed Principal Component Analysis (PCA) for dimensionality reduction of handwriting data, followed by training and evaluation of ten diverse AI models, including logistic regression, Naïve Bayes, random forest, AdaBoost, Support Vector Machine (SVM), and neural network. Model performance was assessed using accuracy, sensitivity, specificity, F1-score, and ROC-AUC metrics. The DARWIN dataset, comprising handwriting samples from 174 participants (89 AD patients, 85 healthy controls) was used for validation. RESULTS: The Neural Network classifier achieved an accuracy of 91% with a 95% CI ranging from 0.79-0.97 and an AUC of 94%, on the test set after identifying the most significant features for AD prediction. These results surpass current clinical diagnostic tools, which typically achieve around 81% accuracy. SMART-Pred's performance aligns with recent AI advancements in AD prediction, such as the Cambridge scientists' AI tool achieving 82% accuracy in identifying AD progression within three years using cognitive tests and MRI scans. The variables "air_time" and "paper_time" consistently emerged as critical predictors for AD across all ten AI models, highlighting their potential importance in early detection and risk assessment. To augment transparency and interpretability, we incorporated the principles of explainable AI, specifically using SHapley Additive exPlanations (SHAP) values, a state-of-the-art method to emphasize the features responsible for our model's efficacy. CONCLUSIONS: SMART-Pred offers non-invasive, cost-effective, and efficient AD prediction, demonstrating the transformative potential of AI in healthcare. While clinical validation is necessary to confirm the practical applicability of the identified key variables, this study contributes to the growing body of research on AI-assisted AD diagnosis and may lead to improved patient outcomes through early detection and intervention.