Department of Radiology, Seoul National University Bundang Hospital, 82 Gumi-ro, 173 Beon-gil, Bundang-gu, Seongnam-si, Gyeonggi-do, 13620, South Korea.
Department of Orthopedic Surgery, Seoul National University Bundang Hospital, 82 Gumi-ro, 173 Beon-gil, Bundang-gu, Seongnam-si, Gyeonggi-do, 13620, South Korea.
Eur Radiol. 2021 Dec;31(12):9408-9417. doi: 10.1007/s00330-021-08034-1. Epub 2021 May 20.
To develop a deep learning algorithm capable of evaluating subscapularis tendon (SSC) tears based on axillary lateral shoulder radiography.
A total of 2,779 axillary lateral shoulder radiographs (performed between February 2010 and December 2018) and the patients' corresponding clinical information (age, sex, dominant side, history of trauma, and degree of pain) were used to develop the deep learning algorithm. The radiographs were labeled based on arthroscopic findings, with the output being the probability of an SSC tear exceeding 50% of the tendon's thickness. The algorithm's performance was evaluated by determining the area under the receiver operating characteristic curve (AUC), sensitivity, specificity, negative predictive value (NPV), and negative likelihood ratio (LR-) at a predefined high-sensitivity cutoff point. Two different test sets were used, with radiographs obtained between January and December 2019; Test Set 1 used arthroscopic findings as the reference standard (n = 340), whereas Test Set 2 used MRI findings as the reference standard (n = 627).
The AUCs were 0.83 (95% confidence interval, 0.79-0.88) and 0.82 (95% confidence interval, 0.79-0.86) for Test Sets 1 and 2, respectively. At the high-sensitivity cutoff point, the sensitivity, NPV, and LR- were 91.4%, 90.4%, and 0.21 in Test Set 1, and 90.2%, 89.5%, and 0.21 in Test Set 2, respectively. Gradient-weighted Class Activation Mapping identified the subscapularis insertion site at the lesser tuberosity as the most sensitive region.
Our deep learning algorithm is capable of assessing SSC tears based on changes at the lesser tuberosity on axillary lateral radiographs with moderate accuracy.
• We have developed a deep learning algorithm capable of assessing SSC tears based on changes at the lesser tuberosity on axillary lateral radiographs and previous clinical data with moderate accuracy. • Our deep learning algorithm could be used as an objective method to initially assess SSC integrity and to identify those who would and would not benefit from further investigation or treatment.
开发一种基于腋侧肩部 X 线摄影的深度学习算法，以评估肩胛下肌腱（SSC）撕裂。
共纳入 2779 例腋侧肩部 X 线摄影（2010 年 2 月至 2018 年 12 月期间拍摄）以及患者的相应临床信息（年龄、性别、优势侧、创伤史和疼痛程度），用于开发深度学习算法。根据关节镜检查结果对 X 线片进行标记，输出结果为 SSC 撕裂超过肌腱厚度 50%的概率。通过确定接收者操作特征曲线（AUC）下面积、敏感度、特异度、阴性预测值（NPV）和阴性似然比（LR-），评估算法在预定高敏感度截止点的性能。使用两种不同的测试集，X 线片分别于 2019 年 1 月至 12 月间拍摄；测试集 1 将关节镜检查结果作为参考标准（n=340），而测试集 2 将 MRI 检查结果作为参考标准（n=627）。
测试集 1 和 2 的 AUC 分别为 0.83（95%置信区间，0.79-0.88）和 0.82（95%置信区间，0.79-0.86）。在高敏感度截止点，测试集 1 的敏感度、NPV 和 LR-分别为 91.4%、90.4%和 0.21，测试集 2 分别为 90.2%、89.5%和 0.21。梯度加权类激活映射确定小结节在较小的突上的肩胛下肌止点是最敏感的区域。
我们的深度学习算法能够基于腋侧 X 线摄影中小结节的变化，以中等准确度评估 SSC 撕裂。
• 我们开发了一种深度学习算法，能够基于腋侧 X 线摄影中小结节的变化和既往临床数据，以中等准确度评估 SSC 撕裂。• 我们的深度学习算法可以作为一种客观方法，用于初步评估 SSC 的完整性，并确定哪些患者将受益于进一步检查或治疗，哪些患者不会受益。
