Database solutions suitable for XAI cardiovascular disease predictions

Sammanfattning

This thesis investigates the integration of Explainable Artificial Intelligence (XAI) techniques into heart disease prediction using database solutions. The primary focus is on comparing the performance and feasibility of local MySQL databases and cloud-based Azure MySQL databases for handling large datasets and providing reliable predictions. The study implements R Shiny for the user interface, ensuring transparency and user-friendliness. Benchmark tests reveal that Azure MySQL outperforms local MySQL in terms of efficiency and scalability. The findings suggest that cloud-based solutions are more suitable for large-scale healthcare applications, offering enhanced performance and better handling of concurrent operations. This project contributes to the field of personalized medicine by demonstrating the practical application of advanced AI and database technologies in predicting heart disease.