Abstract:
Cardiovascular disease (CVD) is a leading cause of death, especially among diabetes
patients, due to metabolic and lifestyle factors. This study aims to predict
cardiovascular disease risk among Bangladeshi diabetes patients using machine
learning and explainable AI, focusing on the role of Mediterranean diet adherence.
Cardiovascular disease (CVD) is a leading cause of morbidity and mortality in diabetes
patients, and early prediction can significantly improve health outcomes. A predictive
model based on machine learning algorithms, particularly LightGBM , was developed
and evaluated, achieving the highest accuracy of 99.37%. The model was further
enhanced with explainable AI techniques to ensure transparency and interpretability
of predictions. The dataset utilized clinical, demographic, and lifestyle data, including
factors such as triglycerides, sleep patterns, smoking habits, and Mediterranean diet
adherence. Results revealed that smoking, sleep hours, and diet adherence were the
most influential factors in predicting cardiovascular risk. The model demonstrated
strong performance across various evaluation metrics, including precision, recall, and
F1-score, further validating its effectiveness. This research underscores the potential
of machine learning to transform healthcare by providing early diagnosis tools,
allowing for personalized interventions. The model can assist healthcare professionals
in identifying at-risk individuals and reducing the burden of cardiovascular diseases
in Bangladesh and similar regions, ultimately improving patient outcomes through
early detection and targeted-prevention-strategies.
