Department of Radiology, The 989th Hospital of the PLA Joint Logistics Support Force, Luoyang, Henan Province, China.
Department of Gastroenterology and Endocrinology, The 989th Hospital of the PLA Joint Logistics Support Force, Luoyang, Henan Province, China.
Injury. 2024 Dec;55(12):111997. doi: 10.1016/j.injury.2024.111997. Epub 2024 Oct 31.
Missed fractures are the most common radiologic error in clinical practice, and erroneous classification could lead to inappropriate treatment and unfavorable prognosis. Here, we developed a fully automated deep learning model to detect and classify femoral neck fractures using plain radiographs, and evaluated its utility for diagnostic assistance and physician training.
1527 plain pelvic and hip radiographs obtained between April 2014 and July 2023 at our Hospital were selected for the model training and evaluation. Faster R-CNN was used to locate the femoral neck. DenseNet-121 was used for Garden classification of the femoral neck fracture, while an additional segmentation method used to visualize the probable fracture area. The model was assessed by the area under the receiver operating characteristic curve (AUC). The accuracy, sensitivity, and specificity for clinicians fracture detection in the diagnostic assistance and physician training experiments were determined.
The accuracy of the model for fracture detection was 94.1 %. The model achieved AUCs of 0.99 for no femoral neck fractures, 0.94 for Garden I/II fractures, and 0.99 for Garden III/IV fractures. In the diagnostic assistance study, the emergency physicians had an average accuracy of 86.33 % unaided and 92.03 % aided, sensitivity of 85.94 % unaided and 91.78 % aided, and specificity of 87.88 % unaided and 93.13 % aided in detecting fractures. In the physician training study, the accuracy, sensitivity, and specificity of the trainees for fracture classification were 81.83 %, 77.28 %, and 84.85 %, respectively, before training, compared with 90.65 %, 88.31 %, and 92.21 %, respectively, after training.
The model represents a valuable tool for physicians to better visualize fractures and improve training outcomes, indicating deep learning algorithms as a promising approach to improve clinical practice and medical education.
漏诊骨折是临床实践中最常见的放射学错误，如果分类错误可能导致治疗不当和预后不良。在这里，我们开发了一种完全自动化的深度学习模型，用于使用骨盆和髋关节平片检测和分类股骨颈骨折，并评估其在诊断辅助和医师培训方面的效用。
从 2014 年 4 月至 2023 年 7 月在我院获得的 1527 张骨盆和髋关节平片被用于模型训练和评估。使用 Faster R-CNN 定位股骨颈，使用 DenseNet-121 对股骨颈骨折进行 Garden 分类，同时使用额外的分割方法可视化可能的骨折区域。使用受试者工作特征曲线下面积（AUC）评估模型。在诊断辅助和医师培训实验中，确定了临床医生检测骨折的准确性、敏感性和特异性。
该模型对骨折检测的准确率为 94.1%。对于无股骨颈骨折，该模型的 AUC 为 0.99；对于 Garden I/II 骨折，AUC 为 0.94；对于 Garden III/IV 骨折，AUC 为 0.99。在诊断辅助研究中，急诊医生在未辅助和辅助条件下的平均准确率分别为 86.33%和 92.03%，敏感性分别为 85.94%和 91.78%，特异性分别为 87.88%和 93.13%。在医师培训研究中，受训者在骨折分类方面的准确性、敏感性和特异性分别为 81.83%、77.28%和 84.85%，培训后分别为 90.65%、88.31%和 92.21%。
该模型为医生更好地可视化骨折和提高培训效果提供了有价值的工具，表明深度学习算法是改善临床实践和医学教育的有前途的方法。
