Determinants of developing cardiovascular disease risk with emphasis on type-2 diabetes and predictive modeling...

Area of Science:

• Cardiology and Public Health
• Utilizing machine learning for disease prediction
• Epidemiology of non-communicable diseases

Background:

• Type-2 diabetes is a growing global health concern.
• Cardiovascular diseases (CVD) are a leading cause of mortality worldwide.
• Understanding diabetes-specific CVD risk factors is crucial for prevention.

Purpose of the Study:

• To investigate the influence of type-2 diabetes on cardiovascular disease (CVD) risk.
• To identify key determinants of CVD risk in individuals with hypertension in Bangladesh.
• To develop and evaluate machine learning (ML) models for precise CVD risk prediction.

Main Methods:

• Analysis of data from the Bangladesh Demographic and Health Surveys (2011, 2017-2018).
• Application and comparison of eight machine learning algorithms (SVM, Logistic Regression, Decision Tree, Random Forest, Naïve Bayes, KNN, LightGBM, XGBoost).
• Evaluation of model performance using six metrics, including accuracy and AUC.

Main Results:

• Older age groups (35-54, 55-69, ≥70 years) showed significantly higher CVD risk.
• Individuals with higher socioeconomic status ('rich') and normal to obese weight had increased CVD risk.
• The predictive models demonstrated high performance, with 75.21% accuracy and 80.79% AUC; Random Forest showed 76.96% specificity.

Conclusions:

• Age, socioeconomic status, and weight are significant determinants of CVD risk in individuals with type-2 diabetes.
• Machine learning models offer a promising approach for early CVD risk assessment and targeted interventions.
• Findings support the need for public health strategies addressing lifestyle changes and improving healthcare access.