Predictive Modelling For Diabetes Mellitus And Cardiovascular Disease Detection Using Artificial Neural Network

Diabetes mellitus and cardiovascular disease (CVD) are two of the most prevalent and interconnected health conditions worldwide, posing significant challenges to healthcare systems and patient well-being. Predictive modelling has emerged as a valuable tool for early detection and risk assessment, facilitating proactive interventions. This study employs artificial neural network (ANN) to develop a predictive model for identifying individuals at risk of concurrent diabetes and CVD. A comprehensive dataset containing 14 attributes as independent and 2 dependent factors is utilized to train, test and validate ANN. The model leverages its capacity to learn complex patterns and relationships within the data, enhancing predictive accuracy. Key input features include gender, age, education, currentSmoker, cigsPerDay, BPMeds, prevalentStroke, prevalentHyp, totChol, sysBP, diaBP, BMI, heartrate and glucose, and output features include 10yearCHD and diabetes. ANN is fine-tuned through iterative optimization techniques to achieve optimal performance. Results demonstrate the efficacy of the ANN-based predictive model in accurately identifying individuals at risk of diabetes and CVD, with promising sensitivity and specificity values. Early detection of these comorbid conditions enables timely interventions, personalized treatment plans, and improved patient outcomes. This research underscores the potential of artificial neural networks as a powerful tool in healthcare for enhancing disease prevention and management strategies