Individualized decision making in on-scene resuscitation time for out-of-hospital cardiac arrest using reinforcement learning.
npj Digital Medicine · 2024-10-09 · Method & validation
Abstract
On-scene resuscitation time is associated with out-of-hospital cardiac arrest (OHCA) outcomes. We developed and validated reinforcement learning models for individualized on-scene resuscitation times, leveraging nationwide Korean data. Adult OHCA patients with a medical cause of arrest were included (N = 73,905). The optimal policy was derived from conservative Q-learning to maximize survival. The on-scene return of spontaneous circulation hazard rates estimated from the Random Survival Forest were used as intermediate rewards to handle sparse rewards, while patients' historical survival was reflected in the terminal rewards. The optimal policy increased the survival to hospital discharge rate from 9.6% to 12.5% (95% CI: 12.2-12.8) and the good neurological recovery rate from 5.4% to 7.5% (95% CI: 7.3-7.7). The recommended maximum on-scene resuscitation times for patients demonstrated a bimodal distribution, varying with patient, emergency medical services, and OHCA characteristics. Our survival analysis-based approach generates explainable rewards, reducing subjectivity in reinforcement learning.
Abstract © the authors or publisher, via PubMed. Reuse and removal
Tags
Mortality & survival · Quality of life & function · Decision support tools
Related articles
- Identifying individuals satisfying the termination of resuscitation rule but having potential to achieve favourable neurological outcome following out-of-hospital cardiac arrest. (Resuscitation, 2023)
- North American validation of the Bokutoh criteria for withholding professional resuscitation in non-traumatic out-of-hospital cardiac arrest. (Resuscitation, 2019)
- A New Rule for Terminating Resuscitation of Out-of-Hospital Cardiac Arrest Patients in Japan: A Prospective Study. (J Emerg Med, 2017)
- The scene time interval and basic life support termination of resuscitation rule in adult out-of-hospital cardiac arrest. (J Korean Med Sci, 2015)
- Beyond the chain of survival: a scoping review of artificial intelligence applications in cardiac arrest. (World J Emerg Med, 2026)
- Pre-hospital neurological risk stratification at return of spontaneous circulation (ROSC) after out-of-hospital cardiac arrest - the Pre-MIRACLE2 score. (Resuscitation, 2026)