Chronic Kidney Disease Stage Prediction Using SGD with ElasticNet Regularization: A Streamlit Application

Chronic Kidney Disease (CKD) is a progressive condition marked by a gradual decline in kidney function. Predicting the stages of CKD accurately is essential for timely interventions and effective disease management. In this study, we present a predictive model that employs Stochastic Gradient Descent (SGD) with ElasticNet regularization to classify CKD stages. ElasticNet regularization, which combines the benefits of both Ll and L2 regularization, enhances model performance by preventing overfitting and addressing multicollinearity. We evaluate our model using a dataset that includes various clinical and biochemical features. Our findings indicate that the proposed method achieves high accuracy and robustness in predicting CKD stages, surpassing traditional regression methods. This model demonstrates significant potential in assisting healthcare professionals with early diagnosis and personalized treatment planning for CKD patients. The SGD with ElasticNet regular-ization model achieved an accuracy of 85% in predicting CKD stages, the highest among the methods tested. This model has been chosen and deployed as the ultimate predictive system, accessible through a web application built using the Streamlit library.