A SMART HEALTHCARE MONITORING SYSTEM UTILISING MACHINE LEARNING AND INTELLIGENT IOT TECHNOLOGY

Abstract

Smart healthcare monitoring systems integrate Internet of Things (IoT) technologies and machine learning to support continuous health monitoring and automated classification of healthcare conditions. This study proposes an intelligent IoT-based healthcare monitoring approach using the IOT-ONN model for health-condition classification. The proposed approach is evaluated using four healthcare datasets obtained from the UCI Machine Learning Repository, namely SPECT Heart, Liver Disorders, Breast Cancer Wisconsin, and Thyroid Disease, covering both binary and multiclass classification problems. The datasets are processed using Python, and the performance of the proposed IOT-ONN model is assessed using ten-fold cross-validation. Its classification performance is compared with Artificial Neural Network (ANN), Naive Bayes, K-Nearest Neighbour, Support Vector Machine, and Decision Tree classifiers. The results demonstrate that IOT-ONN provides improved performance over the conventional ANN model across all selected datasets and achieves competitive classification performance compared with other established machine learning algorithms. The findings indicate that integrating IoT-based healthcare monitoring with intelligent learning models can provide an effective framework for automated health-condition classification and support the development of smart healthcare systems.