法医学杂志 ›› 2023, Vol. 39 ›› Issue (4): 343-349.DOI: 10.12116/j.issn.1004-5619.2023.230308

所属专题: 法医临床鉴定理论与技术专题

• 专题 • 上一篇    下一篇

基于YOLOv3算法的肋骨骨折诊断模型的构建及应用

白洁1(), 孙晶2, 程晓光2(), 刘凡1, 刘华1, 王旭3   

  1. 1.北京市公安局,北京 100192
    2.首都医科大学附属北京积水潭医院,北京 100035
    3.中国政法大学 证据科学教育部重点实验室,北京 100088
  • 收稿日期:2023-03-20 发布日期:2023-10-10 出版日期:2023-08-25
  • 通讯作者: 程晓光
  • 作者简介:程晓光,男,教授,主要从事肌骨影像学研究;E-mail:xiao65@263.net
    白洁(1976—),女,主任法医师,主要从事法医临床鉴定;E-mail:coubai@163.com
  • 基金资助:
    北京市公安局技术研究计划资助项目;北京市医院管理中心临床医学发展专项(ZYLX202107)

Construction and Application of Rib Fracture Diagnosis Model Based on YOLOv3 Algorithm

Jie BAI1(), Jing SUN2, Xiao-guang CHENG2(), Fan LIU1, Hua LIU1, Xu WANG3   

  1. 1.Beijing Public Security Bureau, Beijing 100192, China
    2.Beijing Jishuitan Hospital Affiliated to Capital Medical University, Beijing 100035, China
    3.Key Laboratory of Evidence Law and Forensic Science, Ministry of Education, China University of Political Science and Law, Beijing 100088, China
  • Received:2023-03-20 Online:2023-10-10 Published:2023-08-25
  • Contact: Xiao-guang CHENG

摘要:

目的 建立基于YOLOv3算法的人工智能辅助肋骨骨折诊断模型并应用于实际案例,探讨该模型在法医临床常见肋骨骨折案例中的应用优势。 方法 收集884例胸部外伤致肋骨骨折患者的CT扫描DICOM格式图像,将其中801例作为训练集和验证集,搭建以YOLOv3算法为基础、Darknet53为骨干网络的肋骨骨折诊断模型,建模后以83例为测试集,计算精确率、召回率、F1分数、阅片时间。将该模型用于一起实际案例的诊断,并与人工诊断进行比较。 结果 使用建立的模型对83例进行测试,模型诊断骨折的精确率为90.5%,召回率为75.4%,F1分数为0.82,阅片时间为每秒4.4张,识别每位患者的数据花费时间平均为21 s,远快于人工阅片。所构建模型对实际案例的识别结果与人工诊断结果一致。 结论 基于YOLOv3算法的肋骨骨折诊断模型能够快速、准确地识别骨折,且操作简便,可在法医临床鉴定中作为辅助诊断技术。

关键词: 法医学, 人工智能, 肋骨骨折, 计算机断层扫描, 诊断, YOLOv3, Darknet53

Abstract:

Objective The artificial intelligence-aided diagnosis model of rib fractures based on YOLOv3 algorithm was established and applied to practical case to explore the application advantages in rib fracture cases in forensic medicine. Methods DICOM format CT images of 884 cases with rib fractures caused by thoracic trauma were collected, and 801 of them were used as training and validation sets. A rib fracture diagnosis model based on YOLOv3 algorithm and Darknet53 as the backbone network was built. After the model was established, 83 cases were taken as the test set, and the precision rate, recall rate, F1-score and radiology interpretation time were calculated. The model was used to diagnose a practical case and compared with manual diagnosis. Results The established model was used to test 83 cases, the fracture precision rate of this model was 90.5%, the recall rate was 75.4%, F1-score was 0.82, the radiology interpretation time was 4.4 images per second and the identification time of each patient’s data was 21 s, much faster than manual diagnosis. The recognition results of the model was consistent with that of the manual diagnosis. Conclusion The rib fracture diagnosis model in practical case based on YOLOv3 algorithm can quickly and accurately identify fractures, and the model is easy to operate. It can be used as an auxiliary diagnostic technique in forensic clinical identification.

Key words: forensic medicine, artificial intelligence (AI), rib fracture, computed tomography (CT), diagnosis, YOLOv3, Darknet53

中图分类号: