Machine Learning-Driven Internet of Medical Things (ML-IoMT)-Based Healthcare Monitoring System

Abstract

In order to forecast nine deadly diseases, including blood pressure, diabetes, hepatitis, and kidney disorders, seven machine learning classification algorithms were utilised in this work: adaptive boosting, Random Forest, Decision Trees, Support Vector Machines, Naïve Bayes, Artificial Neural Networks, and K-Nearest Neighbour. Performance criteria such as Accuracy, Precision, and Recall are employed to evaluate the efficacy of the proposed model. Four measures are used to assess the classifiers' performance: accuracy, precision, recall, and precision. The present healthcare model reaches a minimum accuracy of 82.3% and a maximum accuracy of 95.7% for each condition. Every disease has a minimum precision of 81.4% and a maximum precision of 95.7%, as well as a minimum recall of 64.3% and a maximum recall of 90.3%