A Trustworthy Edge Artificial Intelligence Framework for Smart Healthcare Monitoring Using Internet of Medical Things

Abstract

The rapid growth of the Internet of Medical Things (IoMT) has enabled continuous patient monitoring through wearable sensors and intelligent medical devices. However, cloud-based healthcare systems often suffer from latency, bandwidth limitations, and privacy concerns. This paper proposes a trustworthy edge artificial intelligence framework that integrates lightweight deep learning models with edge computing for real-time health monitoring. The framework employs Convolutional Neural Networks (CNNs) for feature extraction and Bidirectional Long Short-Term Memory (Bi-LSTM) networks for temporal analysis of physiological signals such as ECG, SpO₂, body temperature, and heart rate. Explainable Artificial Intelligence (XAI) techniques provide transparent clinical predictions, while lightweight encryption ensures secure data transmission. Experimental evaluation demonstrates improvements in diagnostic accuracy, response time, and energy efficiency compared with conventional cloud-centric approaches. The proposed framework offers a scalable, secure, and intelligent solution for next-generation smart healthcare systems.