利用深度学习为新生儿视频喉镜插管进行声门开放检测。

IF 2.4 3区医学 Q2 OBSTETRICS & GYNECOLOGY

Journal of Perinatology Pub Date : 2024-11-13 DOI:10.1038/s41372-024-02171-3

Abrar Majeedi, Patrick J Peebles, Yin Li, Ryan M McAdams

{"title":"利用深度学习为新生儿视频喉镜插管进行声门开放检测。","authors":"Abrar Majeedi, Patrick J Peebles, Yin Li, Ryan M McAdams","doi":"10.1038/s41372-024-02171-3","DOIUrl":null,"url":null,"abstract":"Objective: This study aimed to develop an artificial intelligence (AI) method to augment video laryngoscopy (VL) by automating the detection of the glottic opening in neonates, as a step toward future studies on improving intubation outcomes.Study design: A deep learning model, YOLOv8, was trained on 1623 video frames from 84 neonatal intubations to detect the glottic opening and evaluated using 14-fold cross-validation on metrics like precision and recall. Additionally, it was compared with 25 medical providers of varied intubation experience to assess its relative performance.Results: The model demonstrated a precision of 80.8% and a recall of 75.3% in identifying the glottic opening, detecting it 0.31 s faster than the average medical provider. It performed comparably or better than novice and intermediate providers, and slightly slower than experts.Conclusion: AI-powered tools can aid VL by providing real-time guidance, potentially enhancing neonatal intubation safety and efficiency for less experienced users.","PeriodicalId":16690,"journal":{"name":"Journal of Perinatology","volume":" ","pages":""},"PeriodicalIF":2.4000,"publicationDate":"2024-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Glottic opening detection using deep learning for neonatal intubation with video laryngoscopy.\",\"authors\":\"Abrar Majeedi, Patrick J Peebles, Yin Li, Ryan M McAdams\",\"doi\":\"10.1038/s41372-024-02171-3\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Objective: This study aimed to develop an artificial intelligence (AI) method to augment video laryngoscopy (VL) by automating the detection of the glottic opening in neonates, as a step toward future studies on improving intubation outcomes.Study design: A deep learning model, YOLOv8, was trained on 1623 video frames from 84 neonatal intubations to detect the glottic opening and evaluated using 14-fold cross-validation on metrics like precision and recall. Additionally, it was compared with 25 medical providers of varied intubation experience to assess its relative performance.Results: The model demonstrated a precision of 80.8% and a recall of 75.3% in identifying the glottic opening, detecting it 0.31 s faster than the average medical provider. It performed comparably or better than novice and intermediate providers, and slightly slower than experts.Conclusion: AI-powered tools can aid VL by providing real-time guidance, potentially enhancing neonatal intubation safety and efficiency for less experienced users.\",\"PeriodicalId\":16690,\"journal\":{\"name\":\"Journal of Perinatology\",\"volume\":\" \",\"pages\":\"\"},\"PeriodicalIF\":2.4000,\"publicationDate\":\"2024-11-13\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Journal of Perinatology\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://doi.org/10.1038/s41372-024-02171-3\",\"RegionNum\":3,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"OBSTETRICS & GYNECOLOGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Perinatology","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1038/s41372-024-02171-3","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"OBSTETRICS & GYNECOLOGY","Score":null,"Total":0}

引用次数: 0

摘要

研究目的本研究旨在开发一种人工智能（AI）方法，通过自动检测新生儿的声门开放情况来增强视频喉镜检查（VL），作为未来改善插管结果研究的一个步骤：研究设计：在 84 个新生儿插管的 1623 个视频帧上训练了深度学习模型 YOLOv8，以检测声门开放情况，并使用 14 倍交叉验证对精确度和召回率等指标进行了评估。此外，该模型还与 25 位具有不同插管经验的医疗服务提供者进行了比较，以评估其相对性能：结果：该模型在识别声门开放方面的精确度为 80.8%，召回率为 75.3%，比普通医疗服务提供者快 0.31 秒。它的表现与新手和中级医疗服务提供者相当或更好，比专家稍慢：人工智能驱动的工具可以通过提供实时指导来帮助 VL，从而为经验不足的用户提高新生儿插管的安全性和效率。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

Glottic opening detection using deep learning for neonatal intubation with video laryngoscopy.

Objective: This study aimed to develop an artificial intelligence (AI) method to augment video laryngoscopy (VL) by automating the detection of the glottic opening in neonates, as a step toward future studies on improving intubation outcomes.

Study design: A deep learning model, YOLOv8, was trained on 1623 video frames from 84 neonatal intubations to detect the glottic opening and evaluated using 14-fold cross-validation on metrics like precision and recall. Additionally, it was compared with 25 medical providers of varied intubation experience to assess its relative performance.

Results: The model demonstrated a precision of 80.8% and a recall of 75.3% in identifying the glottic opening, detecting it 0.31 s faster than the average medical provider. It performed comparably or better than novice and intermediate providers, and slightly slower than experts.

Conclusion: AI-powered tools can aid VL by providing real-time guidance, potentially enhancing neonatal intubation safety and efficiency for less experienced users.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

Journal of Perinatology 医学-妇产科学

CiteScore

5.40

自引率

6.90%

发文量

284

审稿时长

3-8 weeks

期刊介绍： The Journal of Perinatology provides members of the perinatal/neonatal healthcare team with original information pertinent to improving maternal/fetal and neonatal care. We publish peer-reviewed clinical research articles, state-of-the art reviews, comments, quality improvement reports, and letters to the editor. Articles published in the Journal of Perinatology embrace the full scope of the specialty, including clinical, professional, political, administrative and educational aspects. The Journal also explores legal and ethical issues, neonatal technology and product development. The Journal’s audience includes all those that participate in perinatal/neonatal care, including, but not limited to neonatologists, perinatologists, perinatal epidemiologists, pediatricians and pediatric subspecialists, surgeons, neonatal and perinatal nurses, respiratory therapists, pharmacists, social workers, dieticians, speech and hearing experts, other allied health professionals, as well as subspecialists who participate in patient care including radiologists, laboratory medicine and pathologists.