Machine Learning approaches for energy efficient Mechanisms in IoT System for Medical Domain

Abstract

The Internet of Things (IoT) is gaining a lot of traction in numerous industries thanks to its low-cost, self-contained sensors. The Internet of Things (IoT) devices used in healthcare and medicine create an environment for monitoring patients' medical parameters, such as blood pressure, oxygen levels, heart rate, and temperature, and then taking immediate action as needed. Data from patients' medical records is sent to remote users and medical centres for post-analytics using this method. Medical condition monitoring utilising low-powered biosensor nodes has been proposed using a Wireless Body Area Network (WBAN). However, minimising rising energy consumption and communication costs is a pressing issue. An imbalance in the amount of energy consumed by biosensor nodes has a detrimental influence on remote medical centres and the medical system. Furthermore, the patient's private information is being transmitted across an unsecured network that may be attacked. As a result, protecting patient data from unauthorised access and tampering is a pressing research priority in the field of medicine. While delivering healthcare data in an efficient manner, this study's primary goal is to reduce communication overhead and energy consumption between biosensors while also safeguarding patient medical data from unauthentic and malicious nodes to improve the network. The Internet of Medical Things (IoMT) is a proposed framework for e-healthcare that is both secure and energy-efficient. Medical systems' network throughput is increased by 18 percent, packet loss rate is reduced by 44 percent, end-to-end latency is reduced by 26 percent, energy usage by 29 percent, and link breaches by 48 percent using the suggested framework compared to current methods.