Smart Health: Diabetes Prediction System

Saurabh kumar singh
Computer Science and Engineering Parul University
Vadodara, Gujarat, India 2203051050525@paruluniversity.ac.in
Project Guide-prof.Sujaya Bhattcharjee

Abstract:ucial for effective disease management and improved patient outcomes. This research presents the design and implementation of a Smart Health: Diabetes Prediction System utilizing machine learning techniques. The system leverages various classification algorithms, including Logistic Regression, Decision Tree, Random Forest, Support Vector Machine (SVM),and XGBoost, to determine the most accurate predictive model. Hyperparameter tuning is applied using GridSearchCV to optimize model performance.The methodology involves data collection, preprocessing, feature selection, normalization, model training, and evaluation based on accuracy, precision, recall, and F1-score. A web-based application is developed using Flask and Streamlit, allowing real-time predictions for both patients and healthcare professionals. The model is deployed using Streamlit Cloud for enhanced accessibility.Experimental results demonstrate that the system provides a highly accurate and efficient diabetes prediction tool, enablingproactive healthcare interventions. This study highlights the potential of AI-driven solutions in enhancing the accuracy,accessibility, and efficiency of diabetes diagnosis, contributing to theadvancement of predictive healthcare technologies.
Keywords – Diabetes Prediction, Machine Learning, Healthcare AI, Medical Diagnosis, Predictive Analytics, Classification Algorithms, Feature Selection, Hyperparameter Tuning, Web- based Health System, Cloud Deployment, Early Disease Detection.