Smart Diagnosis of Human Cardiovascular Disease Using Machine Learning and Parametric Optimization Techniques

Predicting diseases related to the heart is a difficult task in the medical world. To identify the cause of cardiac problems requires extensive research. This research study has applied various machine learning models to forecast heart disease based on diverse characteristics. In earlier studies, K-nearest neighbour (KNN), logistic regression (LR), support vector machines (SVM), and random forest are used in the procedure of heart syndrome prediction. The main area of worry evolved into model correctness. Gradient Boosting Classifier is used with hyper parameter adjustment to improve accuracy. The fundamental purpose of hyper parameter adjustment is to obtain precise results. The process's output needs be checked, and the system uses a five fold cross validation technique to do so. The data will be split into five sections using this method. The first portion is utilised as testing data in the first iteration, while the following parts are used as training data in subsequent iterations, which continue until all five sections have been covered. Building a special model that makes use of machine learning algorithms and IoT technologies in order to anticipate the occurrence of heart syndrome is the major goal of this method. The API will use the models indicated above to estimate the occurrence of heart disease when the user provides the input data. If the prediction comes to pass, it will reveal that the patient has cardiac disease. The other method of providing user input to the API is by integrating IoT into the project that can sense the input data directly from the user. This project integrates machine learning with IoT technology.