一种基于分割的新型深度学习模型用于提高舟骨骨折检测
本研究开发了一种基于分割的深度学习模型,用于从手腕平片检测显性和隐匿性舟骨骨折,并与三位临床专家比较。模型敏感度0.86、特异度0.83、AUC 0.92,检测隐匿性骨折58例中24例(41%),专家为10.3%、13.7%、6.8%。
要点
- 基于分割的深度学习模型检测舟骨骨折敏感度0.86、特异度0.83、AUC 0.92。
- 模型检测隐匿性舟骨骨折58例中24例(41%),专家为10.3%–13.7%。
- 模型对显性舟骨骨折识别与专家相似,对隐匿性骨折诊断准确性更高。
核心结论
- 使用基于分割的深度学习模型检测舟骨骨折是可行的,与既往开发的深度学习模型相当;模型识别显性舟骨骨折的表现与一组专家相似,并在检测隐匿性骨折方面表现出更高的诊断准确性。
研究设计与样本
- 研究纳入408例患者、410个手腕和1011张手腕平片,其中718张经MRI或CT证实存在舟骨骨折,58例为隐匿性骨折;图像分为训练集、测试集和隐匿性骨折测试集,由标注舟骨和可能骨折区域进行训练,并与三位临床专家的评估比较。
诊断标准与关键数值
- 深度学习模型敏感度0.86(95%CI 0.75-0.93)、特异度0.83(95%CI 0.64-0.94)、准确率0.85(95%CI 0.76-0.92)、AUC 0.92(95%CI 0.86-0.97);临床专家敏感度范围0.77–0.89、特异度范围0.83–0.97;模型检测隐匿性骨折58例中24例(41%),专家分别为10.3%、13.7%、6.8%。
临床应用与阅片要点
- 原文未据此给出推荐。
局限性
- 原文未报告局限性。
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)。
英文原文(摘要)· 点击展开
PURPOSE: To develop a deep learning model to detect apparent and occult scaphoid fractures from plain wrist radiographs and to compare the model's diagnostic performance with that of a group of experts.
MATERIALS AND METHODS: A dataset comprising 408 patients, 410 wrists, and 1011 radiographs was collected. 718 of these radiographs contained a scaphoid fracture, verified by magnetic resonance imaging or computed tomography scans. 58 of these fractures were occult. The images were divided into training, test, and occult fracture test sets. The images were annotated by marking the scaphoid bone and the possible fracture area. The performance of the developed DL model was compared with the ground truth and the assessments of three clinical experts.
RESULTS: The DL model achieved a sensitivity of 0.86 (95 % CI: 0.75-0.93) and a specificity of 0.83 (0.64-0.94). The model's accuracy was 0.85 (0.76-0.92), and the area under the receiver operating characteristics curve was 0.92 (0.86-0.97). The clinical experts' sensitivity ranged from 0.77 to 0.89, and specificity from 0.83 to 0.97. The DL model detected 24 of 58 (41 %) occult fractures, compared to 10.3 %, 13.7 %, and 6.8 % by the clinical experts.
CONCLUSION: Detecting scaphoid fractures using a segmentation-based DL model is feasible and comparable to previously developed DL models. The model performed similarly to a group of experts in identifying apparent scaphoid fractures and demonstrated higher diagnostic accuracy in detecting occult fractures. The improvement in occult fracture detection could enhance patient care.
引用
- Bützow A, Anttila TT, Haapamäki V, et al. A novel segmentation-based deep learning model for enhanced scaphoid fracture detection. Eur J Radiol. 2025. PMID: 40652603. doi:10.1016/j.ejrad.2025.112309
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。