Patient Health Observation and Analysis With Machine Learning And IoT Based in Realtime Environment

Abstract

Today rapid growth of communication and internet technologies has resulted in a significant enhancement of IoT devices. Even the best hospitals and doctors need to develop more in terms of patient care. In case of long waiting periods, a long term gap between doctor visits, inadequate data collection, and other challenges may create problems to healthcare professionals from giving the best care possible. The patients who are suffering from chronic diseases for them ehealthcare are a daily concern. They require disease management tools not only during their doctor visits but every day. In global pandemic situations like today’s, this automated software with the features of Machine Learning will help patients and doctors to maintain physical distance; doctors can monitor patients and prescribe medication to the respective patient from anywhere. Whenever Doctor is unable to monitor patient then this IoT based Machine Learning model will help patients to provide proper medicine through medical staff available based on the symptoms and reports from the IoT sensors with the Machine Learning (ML) trained data set. Here the results obtained for prediction of diabetes and heart diseases, through various machine learning approaches are shown. The obtained results show that for the Gradient Boost, KNN, Random Forest Based classification approaches classify the diseases with higher accuracy rates than the existing models.