The research presents a high-level framework for predicting cardiovascular disease (CVD) risk using machine learning techniques tailored to the Ethiopian healthcare context. It focuses on addressing the challenges of early CVD detection in resource-limited settings by developing an interpretable model trained on electronic medical record (EMR) data from public hospitals in Addis Ababa. The study integrates ensemble learning methods with explainable artificial intelligence tools, combining data-driven prediction with transparent model interpretation. By incorporating SHAP for feature attribution and a large language model (LLM) for translating model outputs into plain-language explanations, the approach aims to create a clinician-friendly and trustworthy decision-support system that can enhance disease risk assessment and support evidence-based healthcare delivery in Ethiopia.

The research presents a high-level framework for predicting cardiovascular disease (CVD) risk using machine learning techniques tailored to the Ethiopian healthcare context. It focuses on addressing the challenges of early CVD detection in resource-limited settings by developing an interpretable model trained on electronic medical record (EMR) data from public hospitals in Addis Ababa. The study integrates ensemble learning methods with explainable artificial intelligence tools, combining data-driven prediction with transparent model interpretation. By incorporating SHAP for feature attribution and a large language model (LLM) for translating model outputs into plain-language explanations, the approach aims to create a clinician-friendly and trustworthy decision-support system that can enhance disease risk assessment and support evidence-based healthcare delivery in Ethiopia.