Volume 1 - Issue 4, November - December 2025

Diabetes, cardiovascular and cancer are the examples of chronic diseases that continue to create serious pressure on the available healthcare infrastructure throughout the globe and demand high-end equipment to detect the cases of diseases early on, simplify the treatment process, and manage this illness throughout the long-term. Artificial intelligence (AI) has been highly predictive in these domains, yet has not been extensively used in clinical practice due to most of the models not being capable of providing mechanistic information about decision making due to being black-box. This obscurity limits the trust and acceptance of physicians and patients especially in high stakes medical situations. This issue can be addressed by coming up with an Explainable and Expandable AI framework which combines Local Interpretable Model-Agnostic Explanations (LIME) and SHapley Additive exPlanations (SHAP) to decode model predictions. The framework highlights the role of features in the local and global level and is able to model the outcome of treatment and make use of individual therapy recommendations. Results of experimental performance on dataset of diabetes, cardiovascular disease and cancer demonstrate this, with an improvement on the interpretability measures and physician trust scores of 17 and 12 percent respectively and maintain the predictive performance of over 92. To a greater extent, patient-specific treatment simulations were more adherent to clinical guidelines, and more so patient-specific advice was more advantageous by 15 percent in comparison to black-box models. The framework also had good cross-disease applicability which guaranteed that the frameworks were not inconsistent in their interpreting of different clinical areas. It implies that the suggested system facilitates the transparency and stability of AI-based diagnostics, an enormous leap in the direction of the digital accuracy medicine and the increase in the application of AI in daily healthcare decision-making.