Deploying Low-Latency Edge AI in Medical IOT Networks: A Case Study of Secure Real-Time Patient Monitoring Systems

Abstract

This study examined the deployment of low-latency Edge Artificial Intelligence (Edge AI) within Medical Internet of Things (MIoT) networks to enhance secure real-time patient monitoring systems through a quantitative experimental case study design. The research aimed to evaluate system performance across key indicators, including latency, throughput, predictive accuracy, energy consumption, and security overhead, by comparing edge-based and cloud-based architectures under controlled operational conditions. A structured dataset comprising 480 experimental iterations was analyzed, incorporating physiological data streams such as heart rate, oxygen saturation, respiratory rate, and electrocardiographic signals. The findings demonstrated that the edge-based system significantly reduced end-to-end latency, achieving a mean delay of 37.8 milliseconds compared to 109.6 milliseconds in the cloud configuration, representing a reduction of approximately 65.5%. Throughput performance was also improved, with the edge system processing 6.2 MB per minute versus 5.5 MB per minute in the cloud system. Predictive accuracy remained high, reaching 96.4% in the edge environment compared to 94.1% in the cloud setup. Energy efficiency analysis indicated that overall system energy consumption was reduced, with edge devices averaging 2.9 watts compared to 3.7 watts for cloud-based processing. Statistical analysis confirmed that these differences were significant at p < 0.05, with large effect sizes observed for latency and processing time improvements. Additionally, the edge system maintained stable performance under high workload and low bandwidth conditions, demonstrating enhanced scalability and network resilience. Security evaluation revealed that encryption overhead introduced a smaller latency increase in the edge system, further supporting its suitability for real-time applications. These results provided strong quantitative evidence that Edge AI significantly improved responsiveness, efficiency, and reliability in MIoT-based patient monitoring systems, offering a robust framework for optimizing healthcare technology deployment in latency-sensitive environments.