Chronic Kindey Disease Prediction Using Machine Learning

Chronic Kidney Disease (CKD) is a critical health condition requiring early detection to prevent severe complications. This project develops a machine learning-based predictive model for CKD detection using clinical parameters such as specific gravity, blood pressure, hemoglobin levels, diabetes status, and appetite. The workflow involves data preprocessing, missing value handling, categorical encoding, feature selection, and model evaluation. Key machine learning models, including Random Forest Classifier, Adobos, and Gradient Boosting, are implemented to achieve high predictive accuracy. After correlation analysis, eight significant features were selected for model training. A Flask-based web application was developed to allow users to input medical parameters and receive real-time CKD risk predictions. The system offers an automated, efficient, and accessible tool for early diagnosis, assisting healthcare professionals in data-driven decision-making to improve patient outcomes.