An Internet Of Medical Things (Iomt)-Based Model For Predicting Optimal Temperatures And Heart Rates On Hypertemia Patients

With the rising use of computer devices and the increasing application of internet of things, many domain have incorporated it to make life easy. One of such is in the medical field where wearable devices are used to check heart beats, perform medical procedure and so on. This research proposed an Internet of Medical Things (IoMT) model based on the Continual Neural Machine Learning Artificial Neural Network (CNML-ANN). The model was adapted for the prediction of Hyperthemia condition of a patient based on the detection of temperature and heart-rate signals. The results of simulations showed the optimal temperatures of between 36^oC and 42^oC for the Temperature Block (TB) and between 30pulses and 120pulses for the Heart Rate Block (HRB) for a total of 200 training data points have been synthesized. In addition, the numbers of false positives are higher for temperature predictions and zero for heart rate using proposed approach. When compared to the Long Short-Term Memory Artificial Neural Network (LSTM-ANN), the classifications for the temperature symptoms is much better with a mean CA of 86% and 68.6% for CNML-ANN and LSTM-ANN respectively. Also, the classifications for the heart rate symptoms is much better with an estimated mean CA of 98% and 82.5% CNML-ANN and LSTM-ANN respectively. Thus, based on the performance of the developed continual learning predictive classifier system, it holds a promising potential as a candidate IoMT machine learning model for real-time diagnosis of Hyperthemia patients.