Search research articles

关于 JoVE

概览领导团队博客 JoVE 帮助中心

作者

出版流程编辑委员会范围与政策同行评审常见问题投稿

图书馆员

用户评价订阅访问资源图书馆顾问委员会常见问题

研究

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments 存档

教育

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual 教师资源中心教师网站

使用条款与条件

相关概念视频

您也可能阅读

相关文章

通过共同作者、期刊和引用图与本文相关的文章。

排序

Same author

Characterizing the Mutational Landscape of In-Transit Melanoma Metastases.

Pigment cell & melanoma research·2026

Same author

The Relationship Between Neutrophil Extracellular Traps and CD8<sup>+</sup> T Lymphocytes in Cancer: A Comprehensive Review of Current Data.

Cancers·2026

Same author

Cancer risk of glucagon-like peptide-1 receptor agonists for obesity: comparison with bariatric surgery and other weight-loss drugs.

Journal of gastrointestinal surgery : official journal of the Society for Surgery of the Alimentary Tract·2026

Same author

Weight loss and cardiovascular outcomes with incretin-based therapies after metabolic and bariatric surgery: a nationwide US cohort study.

EClinicalMedicine·2026

Same author

Nationwide Analysis of Sociodemographic Disparities in Health-Related Quality of Life of Colon Cancer Survivors.

Cancer medicine·2026

Same author

Feasibility of intraoperative force quantification during robotic paraesophageal hernia repair: a prospective case series using integrated force-sensing technology.

Frontiers in surgery·2026

Search research articles

相关实验视频

Updated: Sep 19, 2025

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Published on: August 30, 2013

一个开放架构的人工智能模型用于CPT编码在乳腺外科手术:开发,验证和前性测试.

Mohamad El Moheb^1,2, Kristin Putman¹, Olivia Sears¹

¹Department of Surgery, University of Virginia School of Medicine, Charlottesville, VA.

Annals of surgery

|June 16, 2025

概括

一个开放架构的人工智能 (AI) 模型准确地从乳腺手术笔记中提取程序代码,提高效率并减少错误. 这种人工智能解决方案为当前编码实践提供了一个透明的替代方案.

关键词:

CPT代码 CPT代码是什么意思人工智能的人工智能是人工智能.自动化自动化自动化自动化账单计费账单计费是指账单上的编码编码编码编码自然语言处理自然语言处理.这是一个开源的开源软件.操作注释操作注释变压器变压器变压器工作流效率的工作流效率.

更多相关视频

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

Digital Hybrid Model Preparation for Virtual Planning of Reconstructive Dentoalveolar Surgical Procedures

Digital Hybrid Model Preparation for Virtual Planning of Reconstructive Dentoalveolar Surgical Procedures

Published on: August 5, 2021

相关实验视频

Last Updated: Sep 19, 2025

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Published on: August 30, 2013

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

Digital Hybrid Model Preparation for Virtual Planning of Reconstructive Dentoalveolar Surgical Procedures

Digital Hybrid Model Preparation for Virtual Planning of Reconstructive Dentoalveolar Surgical Procedures

Published on: August 5, 2021

科学领域:

医疗信息学医疗信息学
医疗保健中的人工智能
手术编码自动化 - 手术编码自动化

背景情况:

目前的操作笔记编码效率低下,容易出现错误,导致财务损失和合规问题.
专有,封闭源代码的人工智能系统由于缺乏透明度和验证而阻碍了采用.
需要一个开放架构的人工智能模型来实现可靠和高效的外科编码.

研究的目的:

开发,验证和前性测试一个开放架构的人工智能 (AI) 模型.
人工智能模型旨在从自由文本乳腺手术操作笔记中提取当前程序术语 (CPT) 代码.
解决当前医疗保健编码中闭源AI系统的局限性.

主要方法:

使用了3259份乳腺外科手术操作笔记 (2017年7月至2023年12月),使用专家验证的CPT代码作为参考标准.
开发并验证了两个基于变压器的AI模型 (345M和3.9B参数),在操作笔记上进行了微调.
使用精度回忆曲线下的面积 (AUPRC) 评估性能,并对268张笔记进行了前性测试.

主要成果:

人工智能模型表现出高性能,在CPT代码提取方面表现优于人类外科医生 (大型模型AUPRC:0.981与外科医生AUPRC:0.937).
交叉验证显示与参考标准 (紧型 AUPRC: 0.976) 有很强的对齐.
未来的测试证实了人工智能模型在提取程序代码方面强大的现实性能.

结论:

一个开放架构的人工智能模型有效地自动化了从外科手术笔记中提取CPT代码.
开发的AI提供了一个可扩展,透明和高效的解决方案,以提高外科编码的准确性.
未来的研究将探索AI的潜力,以超越人类编码器的准确性和可靠性.