Health Monitoring in Internet of Medical Things (IoMT) using Machine Learning (ML) Approaches

In today’s world, the Internet of Medical Things (IoMT) is a term that is catching the attention of many researchers. IoMT is used to exchange data between IoT enabled medical sensors and other smart devices for the purpose of healthcare. In this paper, existing machine learning (ML) techniques and models are used to monitor and predict the health condition of a patient by applying these techniques on the sensor data. Apart from collecting some data from the medical sensors used in a single person, IoT traffic of the medical sensors is generated using the IoT-Flock traffic generator. Extraction and selection of the features from the IoMT traffic is done using pcap-processor, WEKA and python script and then some machine learning algorithms like Naive Bayes, Linear Regression, K-Nearest Neighbour and K-Means Clustering are applied to obtain results for the IoMT sensor traffic received from those medical sensors taken into consideration. The processed data thus obtained is used with a time series prediction model such as ARIMA/GARCH to obtain predicted patterns from the selected features of the traffic. The accuracy obtained using this approach is deduced in terms of Mean Square Error (MSE), Mean Absolute Error (MAE) and Root Mean Squared Error(RMSE) of the selected features. In this paper, we have applied existing ML techniques to process the IoMT traffic and make predictions based on it which can help in identifying and predicting the health condition of a patient using IoMT sensors.