Kim Taegyun, Suh Gil Joon, Kim Kyung Su, Kim Hayoung, Park Heesu, Kwon Woon Yong, Park Jaeheung, Sim Jaehoon, Hur Sungmoon, Lee Jung Chan, Shin Dong Ah, Cho Woo Sang, Kim Byung Jun, Kwon Soyoon, Lee Ye Ji
Department of Emergency Medicine, Seoul National University College of Medicine, 103 Daehak-ro, Jongno-gu, Seoul 03080, Republic of Korea; Department of Emergency Medicine, Seoul National University Hospital, 101 Daehak-ro, Jongno-gu, Seoul 03080, Republic of Korea; Research Center for Disaster Medicine, Seoul National University Medical Research Center, 103 Daehak-ro, Jongno-gu, Seoul 03080, Republic of Korea.
Department of Emergency Medicine, Seoul National University Hospital, 101 Daehak-ro, Jongno-gu, Seoul 03080, Republic of Korea.
Resuscitation. 2024 Sep;202:110354. doi: 10.1016/j.resuscitation.2024.110354. Epub 2024 Aug 8.
We evaluated whether an artificial intelligence (AI)-driven robot cardiopulmonary resuscitation (CPR) could improve hemodynamic parameters and clinical outcomes.
We developed an AI-driven CPR robot which utilizes an integrated feedback system with an AI model predicting carotid blood flow (CBF). Twelve pigs were assigned to the AI robot group (n = 6) and the LUCAS 3 group (n = 6). They underwent 6 min of CPR after 7 min of ventricular fibrillation. In the AI robot group, the robot explored for the optimal compression position, depth and rate during the first 270-second period, and continued CPR with the optimal setup during the next 90-second period and beyond. The primary outcome was CBF during the last 90-second period. The secondary outcomes were coronary perfusion pressure (CPP), end-tidal carbon dioxide level (ETCO) and return of spontaneous circulation (ROSC).
The AI model's prediction performance was excellent (Pearson correlation coefficient = 0.98). CBF did not differ between the two groups [estimate and standard error (SE), -23.210 ± 20.193, P = 0.250]. CPP, ETCO level and rate of ROSC also did not show difference [estimate and SE, -0.214 ± 7.245, P = 0.976 for CPP; estimate and SE, 1.745 ± 3.199, P = 0.585 for ETCO; 5/6 (83.3%) vs. 4/6 (66.7%), P = 1.000 for ROSC).
This study provides proof of concept that an AI-driven CPR robot in porcine cardiac arrest is feasible. Compared to a LUCAS 3, an AI-driven CPR robot produced comparable hemodynamic and clinical outcomes.
我们评估了人工智能（AI）驱动的机器人心肺复苏（CPR）是否能改善血流动力学参数和临床结局。
我们开发了一种AI驱动的CPR机器人，该机器人利用一个集成反馈系统和一个预测颈动脉血流（CBF）的AI模型。将12头猪分为AI机器人组（n = 6）和LUCAS 3组（n = 6）。在室颤7分钟后，它们接受了6分钟的CPR。在AI机器人组中，机器人在前270秒内探索最佳按压位置、深度和速率，并在接下来的90秒及以后以最佳设置继续进行CPR。主要结局是最后90秒内的CBF。次要结局是冠状动脉灌注压（CPP）、呼气末二氧化碳水平（ETCO）和自主循环恢复（ROSC）。
AI模型的预测性能极佳（Pearson相关系数 = 0.98）。两组之间的CBF无差异[估计值和标准误差（SE），-23.210 ± 20.193，P = 0.250]。CPP、ETCO水平和ROSC发生率也无差异[估计值和SE，-0.214 ± 7.245，CPP的P = 0.976；估计值和SE，1.745 ± 3.199，ETCO的P = 0.585；ROSC为5/6（83.3%）对4/6（66.7%），P =
