Published Paper Details:

This traditional approach, which seems reasonable at first look but may really result in biases, unforeseen errors, and higher expenses, may endanger patients' QoS (Quality-of-Service). Almost all of the hospitals in Bangladesh do not give exercise bikes since there aren't any intelligent technologies that are both scalable and simple to install. The main goal of this project is to help hospitals treat severely sick patients by providing a feedback system that works in as it happens. we provide a standardized architecture, associated nomenclature, and a categorization model for evaluating the critical patient's health status. Predicting patients' total fitness using machine learning (ML) is the fundamental principle of this work. Our data and ML models may be saved and accessed using IBM Watson Studio and the IBM Cloud. Naive Bayes, Logistic Regression, K-Neighbors, Decision Tree, Random Forest, Gradient Boosting, and MLP were the Base Predictors used by our machine learning models. To increase the model's accuracy, the bagging approach of ensemble learning was used. Collectively, ensemble learning techniques include bagging ridge, bagging support vector machines, bagging additional trees, and bagging random forest.