人工智能辅助气管插管:诊断准确性前瞻性观察研究
本研究首次在人体中评估larynGuide™软件辅助气管插管的准确性,该软件与C-MAC®视频喉镜集成。前瞻性纳入132例择期全麻气管插管成人患者,最终分析110例。在108例气管导管正确置入的患者中,larynGuide识别出102例,敏感度0.94(95%CI 0.88-0.98);软件误分类6例(2例假阴性,4例无反馈)。
要点
- larynGuide实时识别气管导管正确置入的敏感度为0.94(95%CI 0.88-0.98)。
- 软件特异性有限,且视频喉镜图像质量(起雾、视野差)影响性能。
- 首次人体研究证实AI辅助实时气管插管的可行性。
核心结论
- larynGuide™辅助实时气管插管在人体中可行,敏感度较高但特异性有限。
研究设计与样本
- 前瞻性、单中心、研究者发起的观察性研究,纳入成人择期全麻气管插管患者132例,最终分析110例。
主要发现
- 108例气管导管正确置入患者中,larynGuide识别102例,敏感度0.94(95%CI 0.88-0.98)。
- 6例软件误分类:2例假阴性(正确置入但建议失败),4例无反馈。
- 2例首次尝试食管置管失败患者中,larynGuide检测出1例。
临床怎么用
- larynGuide可与C-MAC®视频喉镜集成,为气管插管提供实时辅助判断。
局限性
- 视频喉镜图像质量问题(起雾、视野差)影响软件性能。
- 特异性有限,需进一步优化。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
INTRODUCTION: larynGuide™ is a novel assistive software integrated with the C-MAC® videolaryngoscope, which provides guidance during laryngoscopy and advises on tracheal tube position. This first in-human study evaluated the accuracy and reliability of larynGuide compared with the judgment of the airway operator. METHODS: This prospective, single-centre, investigator-initiated, observational study included adult patients undergoing elective surgery requiring general anaesthesia with tracheal intubation. After informed consent and standardised induction of anaesthesia, laryngoscopy and tracheal intubation were performed with a C-MAC® videolaryngoscope with a Macintosh blade by a board-certified anaesthetist. larynGuide ran on a second screen, visible only to the study team but blinded to the airway operator. After tracheal intubation attempts, the airway operator confirmed tracheal tube placement visually and with capnography. The primary outcome was the real-time accuracy of larynGuide in identifying correct tracheal tube placement. RESULTS: We enrolled 132 patients, of whom 110 were analysed. Of 108 patients with correctly placed tracheal tubes, larynGuide identified 102 (sensitivity 0.94, 95%CI 0.88-0.98). In six patients, the software misclassified tracheal tube position: two false negatives (i.e. the software advised a failed tracheal intubation despite correct placement); and four patients with no feedback. Among two patients with unsuccessful tracheal intubation due to oesophageal tube placement at the first attempt, larynGuide detected one. DISCUSSION: This first in-human study has established the feasibility of AI-guided real-time tracheal intubation using larynGuide. The software showed promising sensitivity, while specificity was limited. Videolaryngoscopy image quality issues, including fogging and poor visibility, impaired the performance of the software. larynGuide™ is a new computer program that works with a special camera tool called the C‐MAC® videolaryngoscope. It helps doctors see the throat more clearly and gives advice on where to place a breathing tube. This study tested how accurate and reliable larynGuide was compared to the doctor's own judgment. The study included adults having planned surgery that needed general anaesthesia and a breathing tube. After patients gave permission, a trained doctor used the C‐MAC® videolaryngoscope to place the breathing tube. The larynGuide software ran on a separate screen that only the research team could see. The doctor did not see the software's advice. After putting in the breathing tube, the doctor checked that it was in the right place using sight a test called capnography that measures carbon dioxide in the patient's breath. The main goal was to see how often larynGuide correctly identified the right tube placement. In total, 132 patients joined the study, and 110 were included in the final results. Of the 108 patients who had their breathing tubes correctly placed, larynGuide also showed correct placement in 102 of them. This means the software was right about 94% of the time. In six cases, the software was wrong or did not give an answer. In two patients where the tube went into the wrong place (the food pipe instead of the windpipe), larynGuide identified one of them. This was the first study to test using larynGuide in real people. The software worked well most of the time and showed that computer help during breathing tube placement is possible. However, it did make some mistakes, especially when the camera view was blocked or blurry.
引用
- Fuchs A, Raeber A, Lippuner R, et al. Artificial intelligence-assisted tracheal intubation in humans: a prospective observational study of diagnostic accuracy. Anaesthesia. 2025. PMID: 41243690. doi:10.1111/anae.70063
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。