Optimization of ML Algorithms in CAD Diagnosis Using Grid Search CV

Amongst the world's main causes of death coronary artery disease plays a major role. As an accurate and timely diagnosis is essential for effective treatment many researches focus on the early identification of the disease. The current approach uses machine learning in making the prediction of cardiac illness. The goal is to design an optimized machine learning framework that assists the healthcare system in the coronary artery disease diagnosis. The dataset used here consists of 11 features and 2300 records including the diseased and non-diseased records. The kNN, SVM, Random forest, Gradient Boosting and Decision Tree algorithms are applied on the classification process. The parameters of these models are optimized using the GridSearch optimizer and the models with the optimized parameters are created and tested with the testing data. The optimized Gradient Boosting model with an accuracy of 95% proves the best among the other models in predicting the coronary atherosclerosis.