Secure and Reliable ML-based Disease Detection for a Medical Wireless Body Sensor Networks

The recent development of the Internet of Things (IoT) has enabled a significant technology that aids quick healthcare solutions through the use of smart wearables sensors. Indeed, undesirable events and network threats can appear in any physiological recording in Wireless Body Sensor Networks (WBSN), leading to a misdiagnosis. These events and threats are recognizable by experienced medical staff, thereby it is necessary to identify them before making any diagnosis. In this paper, a secure and energy efficient approach is proposed. For disease detection, our research provide insight into several physiological signals, including the ElectroCardioGram (ECG), ElectroMyoGram (EMG), and Blood Pressure (BP), where the security is achieved by the application of the Advanced Encryption Symmetric (AES) and the Secure Hash Algorithm (SHA). Similarly, to obtain a reasonable range of reliability, a classification procedure based on supervised Machine Learning (ML) techniques is used. The simulation results proved the accuracy and sensitivity of the system by 97% and 92%, respectively by enhancing a high level of security. Moreover, a suitable prototype is developed for medical staff to ensure the applicability of our proposal.