A Hybrid IoT-ML Based Electromagnetic Compatibility Monitoring and Prediction System for Biomedical Devices

The increasing of electromagnetic radiation from wireless electronic devices has raised concerns regarding electromagnetic compatibility (EMC) and the potential for electromagnetic interference (EMI) in healthcare environments especially for critical medical equipment. To address this challenge, regular inspection is required. We propose an innovative solution integrating Internet of Things (IoT) technology and machine learning (ML) algorithms for monitoring and predicting EMC levels for medical equipment. This research focuses on the development and implementation of a wireless sensor network (WSN) capable of capturing realtime data from medical equipment and surrounding environment such as electromagnetic radiation (EMR), temperature, operating voltage and aging of medical devices. This data is integrated into a ML model based on multi linear regression (MLR) to predict EMC and anticipate potential EMI events. The system utilizes a Raspberry Pi 4 for processing data to the edge of network and GSM/GPRS SIM7600 4G HAT for data transmission along with sensors for temperature, EMI, and operating voltage, sending data to cloud for further analysis and visualization. A dashboard is built for remote monitoring via cloud platform. The regression equations generated by ML algorithm, accurately forecast EMC levels based on environmental factors and device characteristics. Alerts are disseminated via email and smartphone notifications to facilitate timely intervention and ensure uninterrupted operation of medical equipment. Our findings warrantee the effectiveness of proposed system for proactively monitoring EMC hazards from biomedical devices and surrounding electronic equipment so as to improve hospital safety for both patients and staff.