Tran Son Quang, Bui Minh Cong, Nguyen Dat Tien, Itthipanichpong Thun, Limskul Danaithep, Thamrongskulsiri Napatpong, Tanpowpong Thanathep
Clinical Sciences Program, Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand.
Faculty of Medicine, Department of Orthopaedics, Cantho University of Medicine and Pharmacy, Can Tho, Vietnam.
JSES Int. 2025 Mar 5;9(3):864-870. doi: 10.1016/j.jseint.2025.02.003. eCollection 2025 May.
The clarity of visualization in shoulder arthroscopy is significantly influenced by intraoperative bleeding. This study aims to develop deep learning models to classify the visual clarity of arthroscopic shoulder images and evaluate their reliability compared to rater assessment.
We retrospectively reviewed videos from 113 shoulder arthroscopies, using 63 patients' videos to create a 3750-image training dataset and 50 patients' videos to evaluate the reliability and agreement of the trained models. Images extracted from the videos were assessed for visual clarity with a 3-grade scale. Subsequently, we implemented transfer learning techniques for the pretrained deep learning models involving DensetNet169, DenseNet201, Xception, InceptionResNetV2, VGG16, and ViT. The reliability and agreement of the trained predictive models compared with raters in classifying the visual clarity of shoulder arthroscopic images were reported with percent agreement and weighted kappa coefficients. Similarly, for quantifying the visual clarity of surgical videos, the reliability and agreement were evaluated through intraclass coefficients, bias, and limits of agreement.
Most models achieved over 90% accuracy in validation step, with InceptionResNetV2, DenseNet169, and ViT exhibiting good percent agreements of 82.3%, 79.2%, and 77.8%, respectively, and their weighted kappa coefficients above 0.8. DenseNet169 model had the highest reliability for evaluating the visual clarity of surgical videos with an intraclass correlation coefficient of 0.9, a bias of 0.027, and the narrowest limits of agreement.
The trained DenseNet169 predictive model was reliable enough to be utilized as an objective measure of the visual clarity of the shoulder arthroscopic field for further research.
肩关节镜检查中的可视化清晰度受术中出血的显著影响。本研究旨在开发深度学习模型，以对关节镜下肩部图像的视觉清晰度进行分类，并与评估者评估相比，评估其可靠性。
我们回顾性分析了113例肩关节镜检查的视频，使用63例患者的视频创建了一个包含3750张图像的训练数据集，并使用50例患者的视频评估训练模型的可靠性和一致性。从视频中提取的图像根据视觉清晰度采用三级量表进行评估。随后，我们对涉及DensetNet169、DenseNet201、Xception、InceptionResNetV2、VGG16和ViT的预训练深度学习模型实施迁移学习技术。报告训练后的预测模型与评估者在对肩关节镜图像的视觉清晰度进行分类时的可靠性和一致性，以百分比一致性和加权kappa系数表示。同样，为了量化手术视频的视觉清晰度，通过组内相关系数、偏差和一致性界限来评估可靠性和一致性。
大多数模型在验证步骤中准确率超过90%，InceptionResNetV2、DenseNet169和ViT分别表现出良好的百分比一致性，为82.3%、79.2%和77.8%，其加权kappa系数高于0.8。DenseNet169模型在评估手术视频的视觉清晰度方面具有最高的可靠性，组内相关系数为0.9，偏差为0.027，一致性界限最窄。
训练后的DenseNet169预测模型足够可靠，可作为肩关节镜视野视觉清晰度的客观测量方法用于进一步研究。
