Prediction of Metabolic Syndrome using Machine Learning Algorithms

Metabolic Syndrome (MetS) refers to a group of metabolic disorders that collectively increase the risk of cardiovascular diseases, type 2 diabetes, and stroke. The parameters include elevated blood pressure, high triglyceride levels, low HDL cholesterol, abdominal obesity, and elevated fasting glucose. The existing systems for diagnosis rely heavily on manual clinical assessments and static threshold-based methods and existing algorithms like Logistic Regression and Basic Decision Trees, which often lack the ability to capture complex interdependencies among risk factors. These methods are time-consuming, susceptible to human error, and offer limited predictive capability, particularly for borderline cases. Accurate early prediction of Metabolic Syndrome can play a key role in preventive healthcare and early intervention. The proposed framework is developed using Random Forest machine learning algorithm trained on clinical data to assess an individual’s risk of developing Metabolic Syndrome. The system utilized structured health datasets and applied data pre-processing techniques before training multiple classifiers. Among the algorithms like Random Forest, Support Vector Machine, Logistic Regression, Long Short-Term Memory, K-Nearest Neighbors and Decision Trees are tested, the Random Forest model demonstrated the highest performance based on accuracy, precision, recall, and F1-score. The proposed model helps in identifying at-risk individuals, offering a cost-effective, non-invasive solution for early screening and diagnosis of Metabolic Syndrome. The final model is deployed as a web application facilitating real time predictions.