Development of an Internet of Things-based Smart Health Monitoring System for COVID-19

The application of machine learning (ML) -based Internet of Things (IoT)- in healthcare during COVID-19 has proven effective in preventing disease spread. This paper proposed an IoT system to remotely detect and monitor suspected COVID-19 cases at home. To develop the system, IoT sensors, a mobile application, and ML algorithms were used for data analysis. The IoT node tracked health parameters, including body temperature, heart rate, blood oxygen saturation level, and blood pressure, as well as age and gender information, then used a user interface to display health status updates and sent them to a cloud server. The decision-making module based on ML algorithms was placed on the server side. The system's efficiency was evaluated using accuracy, precision, recall, and F1 score metrics. Doctors could view and track users' health status through an interface, and Pearson's correlation coefficient was used to determine the correlation between sensor measurements and vital sign devices. The proposed system achieved 99.05% accuracy, 98.66% precision, 99.32% recall, and 98.99% F-measure using the random forest algorithm. The Pearson correlation coefficient also showed a strong correlation in evaluating the sensors. The system could aid in diagnosing and monitoring suspected COVID-19 cases.