人工智能在急诊胸部CT检查诊断肋骨骨折中的临床价值  

The Clinical Value of Artificial Intelligence in the Diagnosis of Rib Fracture by Chest CT in Emergency

作　　者：黄玉罡 蒋海蓉 HUANG Yugang;JIANG Hairong(Department of Radiology,Guangzhou General Hospital,Qianhai Life Insurance,Guangdong 511300,China)

机构地区：[1]前海人寿广州总医院放射诊断科,广东511300

出　　处：《影像技术》2025年第2期70-73,共4页Image Technology

摘　　要：目的:肋骨骨折是常见的急诊病例之一,传统的诊断方法存在一定的局限性。人工智能(AI)作为新兴技术,在医学影像诊断中发挥了重要作用。本研究旨在探讨人工智能在急诊肋骨骨折诊断中的临床价值。方法:收集293例前海人寿广州总医院2023年1月-2024年3月进行急诊胸部CT检查发现肋骨骨折的病人。将所有病人分两次进行阅片,一次使用人工智能系统进行影像分析,一次使用三名高年资医生进行轮流阅片,将人工智能与医生的结果进行灵敏度、阅片时间对比。结果:人工智能诊断灵敏度95.1%,其中错位性骨折灵敏度99.14%,不完全性骨折灵敏度93.59%;人工智能阅片时间(40.23±20.05)s。医生阅片诊断灵敏度78.7%,其中错位性骨折灵敏度94.44%,不完全性骨折灵敏度73.9%;医生阅片时间(124.59±25.87)s。人工智能系统在诊断肋骨骨折方面表现出的灵敏度明显较医生组高,且在阅片时间上明显较医生组短。与传统的诊断方法相比,人工智能系统能够更快速、更准确地识别肋骨骨折,并降低漏诊率。结论:人工智能在急诊胸部CT检查诊断肋骨骨折中具有重要的临床价值,能够提高诊断的准确性和效率,为患者提供更快速的诊断和治疗。Objective:Rib fracture is one of the common emergency cases.The traditional diagnostic methods have some limitations.Artificial intelligence(AI),as an emerging technology,has the potential to play an important role in medical imaging diagnostics.The purpose of this study was to investigate the clinical value of AI in the diagnosis of emergency rib fracture.Methods:293 patients with rib fracture who underwent emergency chest CT examination in Guangzhou General Hospital of Foresea Life Insurance from January 2023 to March 2024 were collected.All patients were divided into two times to read the image,one time using AI system for image analysis,one time using three senior doctors to read the image in turn,and the sensitivity and reading time of the results of AI and doctors were compared.Results:The diagnostic sensitivity of AI was 95.1%,the sensitivity of malposition fracture was 99.14%,the sensitivity of incomplete fracture was 93.59%;AI image reading time was(40.23±20.05)s.The diagnostic sensitivity was 78.7%,including 94.44%for malposition fracture and 73.9%for incomplete fracture.The doctor's reading time was(124.59±25.87)s.The sensitivity of the AI system in the diagnosis of rib fracture was significantly higher than that of the doctor group,and the image reading time was significantly shorter than that of the doctor group.Compared to traditional diagnostic methods,the AI system can identify rib fractures more quickly and accurately,and reduce the rate of missed diagnoses.Conclusion:AI has important clinical value in the diagnosis of rib fracture by chest CT examination in emergency,which can improve the accuracy and efficiency of diagnosis and provide patients with faster diagnosis and treatment.

关 键 词：人工智能 急诊 肋骨骨折 胸部CT 诊断 

分 类 号：R812[医药卫生—放射医学] R816.41[医药卫生—临床医学]