An Efficient LightGBM Approach for Cardiovascular Disease Detection Highlighting Cholesterol and Body Mass Index
Obesity is a health condition that directly impacts an individual's quality of life and overall well-being. Its occurrence is directly associated with lifestyle, ultimately leading to other health conditions like diabetes, hypertension and Cardiovascular Diseases (CVD). In addition to obesity, which is measured by the Body Mass Index (BMI), CVD is also influenced by other factors like age, gender, diabetes, and cholesterol levels. Total cholesterol is another key manageable risk factor for CVD, heart attack or stroke. Heart-related disorders can be managed if detection, risk evaluation and prediction are done at an early stage. This calls for an urgent need for AI-based technologies that could reliably predict the forthcoming consequences to individuals having obesity, high cholesterol or CVD. Feature importance analysis suggests that the BMI and total cholesterol are the most influential markers, followed by pulse pressure, heart rate and glucose level. The findings indicate that the proposed Light Gradient Boosting Machine (GBM) performance model achieved an accuracy of 90%, a recall of 86%, a precision of 95%, an F1 score of 90% and a ROC-AUC of 95%, respectively. These outcomes highlight the crucial role of obesity and cholesterol management in reducing cardiovascular risk. The proposed LightGBM-based framework offers a reliable and interpretable solution for early detection of cardiovascular disease, thereby supporting preventive healthcare and informed clinical decision-making.