Title:
AI Smart Health Risk Assessment

Authors:
Nenavath Chander, P.Siddarda, P.Shiva, T.Aryan, India

Abstract:

AI Smart Health Risk Assessment is a system designed to facilitate the early detection of diabetes and other metabolic disorders using machine learning techniques. Traditional diagnostic approaches typically rely on clinical consultations and laboratory testing, which can be time-consuming and may not be easily accessible to all individuals. To address these limitations, the proposed system employs the Extreme Gradient Boosting (XGBoost) algorithm in combination with a rule-based Clinical Interpretation Engine to provide rapid and reliable health risk predictions. The system analyzes seven key health parameters: age, body mass index (BMI), systolic and diastolic blood pressure, fasting glucose, glycated hemoglobin (HbA1c), and total cholesterol. Based on these inputs, it generates a probability score and classifies the risk into three categories: low, moderate, or high. In addition to prediction, the system offers interpretability by explaining the contribution of each health parameter to the overall risk. It evaluates glucose levels, HbA1c values, blood pressure categories, indicators of metabolic syndrome, and cardiovascular risk factors. Based on this analysis, personalized recommendations are provided, enabling users to make informed decisions regarding medical consultation. All assessment results are stored in a MongoDB database with timestamps, allowing users to monitor their health trends over time through visual representations. Furthermore, the system includes additional functionalities such as BMI calculation, result comparison, medical recommendations, and report generation. Overall, the proposed system demonstrates the potential of artificial intelligence in enhancing preventive healthcare by delivering faster, more accessible, and interpretable risk assessment solutions.

DOI: http://dx.doi.org/10.51505/ijaemr.2026.11231