文献速递运动医学关节外科

基于术中关节镜图像使用深度学习预测关节镜下肩袖修复术后再撕裂

Prediction of Retear After Arthroscopic Rotator Cuff Repair Based on Intraoperative Arthroscopic Images Using Deep Learning.
PubMed 文献 · Am J Sports Med · 2023-08-11

该队列研究(诊断性,证据等级2)回顾性获取580例患者的1394张术中关节镜图像,使用VGG16、DenseNet和Xception三种深度学习架构预测关节镜下肩袖修复术后再撕裂。DenseNet在测试集准确率91%、AUC 0.92,表现最佳。结论认为深度学习算法应用于术中关节镜图像在预测再撕裂方面显示出高准确率。

要点

  • 研究纳入580例患者共1394张术中关节镜图像,非再撕裂组514例1138张,再撕裂组66例256张。
  • 测试集准确率:DenseNet 91%、Xception 87%、VGG16 76%;AUC分别为0.92、0.91、0.83。
  • DenseNet测试集特异度0.93、敏感度0.84;Xception为0.89和0.84;VGG16为0.70和0.80。

核心结论

  • 深度学习算法应用于术中关节镜图像在预测再撕裂发生方面显示出高准确率。

研究设计与样本

  • 队列研究(诊断性),证据等级2;回顾性获取580例患者的1394张术中关节镜图像。
  • 使用VGG16、DenseNet和Xception三种预训练深度学习算法进行迁移学习。
  • 训练集与测试集按8:2划分,采用三折分层验证在训练集上微调超参数。

主要发现

  • 各模型平均验证准确率:VGG16 83%、Xception 89%、DenseNet 91%。
  • 测试集准确率:VGG16 76%、Xception 87%、DenseNet 91%。
  • AUC:DenseNet最高为0.92,VGG16为0.83,Xception为0.91。
  • 测试集特异度和敏感度:DenseNet为0.93和0.84,Xception为0.89和0.84,VGG16为0.70和0.80。

关键数值与推荐

  • 非再撕裂组514例患者共1138张图像,再撕裂组66例患者共256张图像。
  • 修复肌腱完整性通过术后2年内磁共振成像评估。

临床怎么用

  • 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)。

原文

BACKGROUND: It is challenging to predict retear after arthroscopic rotator cuff repair (ARCR). The usefulness of arthroscopic intraoperative images as predictors of the ARCR prognosis has not been analyzed. PURPOSE: To evaluate the usefulness of arthroscopic images for the prediction of retear after ARCR using deep learning (DL) algorithms. STUDY DESIGN: Cohort study (Diagnosis); Level of evidence, 2. METHODS: In total, 1394 arthroscopic intraoperative images were retrospectively obtained from 580 patients. Repaired tendon integrity was evaluated using magnetic resonance imaging performed within 2 years after surgery. Images obtained immediately after ARCR were included. We used 3 DL architectures to predict retear based on arthroscopic images. Three pretrained DL algorithms (VGG16, DenseNet, and Xception) were used for transfer learning. Training and test sets were split into 8:2. Threefold stratified validation was used to fine-tune the hyperparameters using the training data set. The validation results of each fold were evaluated. The performance of each model in the test set was evaluated in terms of accuracy, area under the receiver operating characteristic curve (AUC), F1-score, sensitivity, and specificity. RESULTS: In total, 1138 and 256 arthroscopic images were obtained from 514 patients and 66 patients in the nonretear and retear groups, respectively. The mean validation accuracy of each model was 83% for VGG16, 89% for Xception, and 91% for DenseNet. The accuracy for the test set was 76% for VGG16, 87% for Xception, and 91% for DenseNet. The AUC was highest for DenseNet (0.92); it was 0.83 for VGG16 and 0.91 for Xception. For the test set, the specificity and sensitivity were 0.93 and 0.84 for DenseNet, 0.89 and 0.84 for Xception, and 0.70 and 0.80 for VGG16, respectively. CONCLUSION: The application of DL algorithms to intraoperative arthroscopic images has demonstrated a high level of accuracy in predicting retear occurrences.

来源:Am J Sports Med,查看原文。版权归原作者所有。

引用

  1. Cho SH, Kim YS. Prediction of Retear After Arthroscopic Rotator Cuff Repair Based on Intraoperative Arthroscopic Images Using Deep Learning. Am J Sports Med. 2023. PMID: 37565449. doi:10.1177/03635465231189201

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。