Healthcare Monitoring and Concurrent Medical Service Consultation using Deep Learning and Optimization Techniques in IoMT

There has been substantial expansion of the healthcare monitoring system in hospitals and other healthcare facilities, and many nations around the world are growing increasingly concerned about the availability of transportable health monitoring devices equipped with emerging technologies. In healthcare, the shift from in-person to remote consultations is made easier by the development of Internet of Things technology. This study offers an Internet of Things based healthcare system capable of real-time monitoring of vitals and environmental factors such as temperature and humidity in the patient’s current environment. Five different sensors, including a CO sensor, heart rate sensor body temperature probe, room temp sensor, and a CO2 sensor, are used to gather data from the hospital setting. The error rate of the developed method is always lower than a predetermined limit. A portal allows doctors to keep up with patient data and utilize it to make more informed decisions about treatment. The approach incorporates a cost for skipping over potentially harmful input in order to create a neural network convolution with more efficiency. This research modifies the convolution neural network’s architecture to account for the diversity of data features and the dynamism of data changes over time, with an eye toward eight physiological properties of data. The prediction accuracy of medical evaluation reaches 90.15%, which is higher than some other machine-learning algorithms thanks to the realized CNN optimization algorithm model’s increased prediction effect. The achievement of the system demonstrates that the built model is ideal for health monitoring.