A Novel Artificial Intelligence-Enhanced Digital Network for Prehospital Emergency Support: Community Intervention Study

Abstract

Background: Efficient emergency patient transport systems, which are crucial for delivering timely medical care to individuals in critical situations, face certain challenges. To address this, CONNECT-AI (CONnected Network for EMS Comprehensive Technical-Support using Artificial Intelligence), a novel digital platform, was introduced. This artificial intelligence (AI)-based network provides comprehensive technical support for the real-time sharing of medical information at the prehospital stage.

Objective: This study aimed to evaluate the effectiveness of this system in reducing patient transport delays.

Methods: The CONNECT-AI system provided 3 key AI services to prehospital care providers by collecting real-time patient data from the scene and hospital resource information, such as bed occupancy and the availability of emergency surgeries or procedures, using 5G communication technology and internet of things devices. These services included guidance on first aid, prediction of critically ill patients, and recommendation of the optimal transfer hospital. In addition, the platform offered emergency department medical staff real-time clinical information, including live video of patients during transport to the hospital. This community-based, nonrandomized controlled intervention study was designed to evaluate the effectiveness of the CONNECT-AI system in 2 regions of South Korea, each of which operated an intervention and a control period, each lasting 16 weeks. The impact of the system was assessed based on the proportion of patients experiencing transfer delays.

Results: A total of 14,853 patients transported by public ambulance were finally selected for analysis. Overall, the median transport time was 10 (IQR 7-14) minutes in the intervention group and 9 (IQR 6-13) minutes in the control group. When comparing the incidence of transport time outliers (>75%), which was the primary outcome of this study, the rate was higher in the intervention group in region 1, but significantly reduced in region 2, with the overall outlier rate being higher in the intervention group (27.5%-29.7%, P=.04). However, for patients with fever or respiratory symptoms, the group using the system showed a statistically significant reduction in outlier cases (36.5%-30.1%, P=.01). For patients who received real-time acceptance signals from the hospital, the reduction in the percentage of 75% outliers was statistically significant compared with those without the system (27.5%-19.6%, P=.02). As a result of emergency department treatment, 1.5% of patients in the control group and 1.1% in the intervention group died (P=.14). In the system-guided optimal hospital transfer group, the mortality rate was significantly lower than in the control group (1.54%-0.64%, P=.01).

Conclusions: The present digital emergency medical system platform offers a novel approach to enhancing emergency patient transport by leveraging AI, real-time information sharing, and decision support. While the system demonstrated improvements for certain patient groups facing transfer challenges, further research and modifications are necessary to fully realize its benefits in diverse health care contexts.