设备无关智能手表技术自动检测心脏骤停并警报急救服务:原理验证
本研究为原理验证,从三组收集循环骤停数据:袖带暂时限制手臂血流的志愿者、心脏手术中停循环的患者、食品工业屠宰的家猪,使用三星Watch5和Watch5 Pro采集光电容积脉搏波信号并开发算法检测循环骤停。19名参与者中,31次循环骤停正确识别28次,敏感度90.3%(95%CI 74.2%-98.0%),小时级特异度94.1%(95%CI 71.3%-99.9%)。触发循环骤停一致引起手表视听警报,并通过HartslagNu测试服务器瞬时警报虚拟急救医疗服务。作者认为该技术有改善生存的潜力,但需更高特异度以防止加重急救负担,未来研究应聚焦真实心脏骤停数据的现实验证。
要点
- 在19名参与者中,算法识别循环骤停的敏感度为90.3%(95%CI 74.2%-98.0%),31次骤停中正确识别28次。
- 小时级特异度为94.1%(95%CI 71.3%-99.9%),作者指出需更高特异度以防止加重急救医疗服务负担。
- 触发循环骤停后,智能手表一致产生视听警报,并通过HartslagNu测试服务器瞬时警报虚拟急救医疗服务。
- 数据来自袖带限制手臂血流的志愿者、心脏手术停循环患者和屠宰家猪三组,使用三星Watch5和Watch5 Pro。
- 作者认为该技术有改善生存的潜力,未来研究应聚焦真实心脏骤停数据的现实验证。
适用人群与场景
- 数据来自三组:袖带暂时限制手臂血流的志愿者、心脏手术中停循环的患者、食品工业屠宰的家猪。
- 研究场景为原理验证,使用三星Watch5和Watch5 Pro采集光电容积脉搏波信号。
研究设计与关键结果
- 研究类型为原理验证,共分析19名参与者,使用三星Watch5和Watch5 Pro采集数据并开发算法分析光电容积脉搏波信号。
- 31次循环骤停中正确识别28次,敏感度90.3%(95%CI 74.2%-98.0%),小时级特异度94.1%(95%CI 71.3%-99.9%)。
- 触发循环骤停一致引起智能手表视听警报,并通过HartslagNu测试服务器瞬时警报虚拟急救医疗服务。
证据类型、等级与局限
- 证据类型:原理验证研究。
- 局限:作者指出需更高特异度以防止加重急救医疗服务负担;未来研究应聚焦真实心脏骤停数据的现实验证。
对床旁实践的改变
- 作者认为该技术有改善生存的潜力,但需更高特异度以防止加重急救医疗服务负担。
- 作者建议未来研究应聚焦真实心脏骤停数据的现实验证。
提炼说明
- 本条依据论文摘要整理;摘要未涉及「诊断标准与关键阈值」「处置流程与药物剂量」「红旗征与不能漏的情况」「指南或共识出处」,故未列出这些栏目;全文请点「阅读原文」。
英文原文(摘要)· 点击展开
INTRODUCTION: Out-of-hospital cardiac arrest (OHCA) is a leading cause of mortality. Automated detection could improve survival by reducing delays in first responder activation. This study provides proof-of-principle for a device-independent technology that can (A) distinguish presence versus absence of spontaneous circulation, and (B) reliably alert emergency medical services (EMS).
METHODS: Circulatory arrest data were collected from three groups: (1) volunteers undergoing temporarily restricted blood flow to the arm using a cuff, (2) patients undergoing cardioplegic cardiac arrest for heart surgery, and (3) domestic swine, slaughtered in food industry. Data were collected using Samsung Watch5 and Watch5 Pro. An algorithm was developed to analyze photoplethysmography signals and detect circulatory arrest. Emergency response was tested via the Dutch community first responder network HartslagNu, using their test environment to activate test responders and EMS.
RESULTS: Nineteen participants were analyzed. Across all three groups, 28 of 31 circulatory arrests were correctly identified, sensitivity 90.3% (95% CI: 74.2%-98.0%), and hour-level specificity was 94.1% (95% CI: 71.3%-99.9%). Triggering a circulatory arrest consistently resulted in an audiovisual smartwatch alarm and an instantaneous alert to the virtual EMS at the HartslagNu test server.
CONCLUSION: This study demonstrates the feasibility of detecting circulatory arrest using commercially available smartwatch sensors, achieving high sensitivity and specificity. Additionally, we integrated an automated alerting system with emergency networks to notify first responders. While this technology shows promise to improve survival, higher specificity is needed to prevent overburdening EMS. Future research should focus on real-world validation using actual cardiac arrest data.
参考来源
- van den Beuken WMF, Nideröst B, Goossen SA, et al. Automated cardiac arrest detection and emergency service alerting using device-independent smartwatch technology: proof-of-principle. Resuscitation. 2025. PMID: 40412645. doi:10.1016/j.resuscitation.2025.110657
供执业医师抢救与诊疗参考,不替代临床判断;剂量请结合体重、肝肾功能和本院药品规格复核,以指南原文为准。