Chronic Kidney Disease Detection Using GridSearchCV Cross Validation Method

Abstract

Kidney disease is a serious public health issue that is spreading around the world. According to estimates, 10% of people globally suffer from chronic kidney disease, which is one of the main causes of mortality and disability. Hence, for early identification, prevention, and disease treatment, precise prediction of renal disease is crucial. Overall, renal disease prediction is important for research because it can improve patient outcomes, tailor care, and lead to the creation of fresh preventative and therapeutic approaches. In order to forecast renal illness for this study's GridSearchCV with 10-fold cross-validation, we must first import the required libraries and load the dataset. Secondly, dividing the dataset into features and labels to prepare it for modelling. We created a pipeline that comprises preprocessing procedures and a machine learning algorithm after dividing the data into training and testing sets. Then, using 10-fold cross-validation, fit the GridSearchCV object to the training data after establishing the hyperparameters to search over and using it. Lastly, we forecasted renal illness on the test set using the best estimator discovered by GridSearchCV, and assessed the model's performance using measures like accuracy, precision, and recall.