Detection and classification of supraspinatus pathologies on shoulder magnetic resonance images using a code-free deep learning application

IF 1.5 Q3 ORTHOPEDICS
Onur Engin , Ceren Durmaz Engin , Atilla Hikmet Çilengir , Berna Dirim Mete
{"title":"Detection and classification of supraspinatus pathologies on shoulder magnetic resonance images using a code-free deep learning application","authors":"Onur Engin ,&nbsp;Ceren Durmaz Engin ,&nbsp;Atilla Hikmet Çilengir ,&nbsp;Berna Dirim Mete","doi":"10.1016/j.asmart.2025.04.005","DOIUrl":null,"url":null,"abstract":"<div><h3>Objective</h3><div>To evaluate the performance of a code free deep learning (CFDL) application in diagnosing supraspinatus tendon pathologies on shoulder magnetic resonance imaging (MRI) images.</div></div><div><h3>Design</h3><div>This retrospective cross-sectional study included patients with supraspinatus MRI showing partial or full-thickness tears and tendinosis, with patients having normal findings as the control group. MRI images were processed in the LobeAI application using transfer learning with ResNet-50 V2 for model development. Models were built to differentiate each pathology from normal and full-thickness tears from partial tears.</div></div><div><h3>Results</h3><div>The ML models developed using the LobeAI application demonstrated the ability to differentiate between normal shoulder MRI images and partial tears, full-thickness tears, and tendinosis with sensitivities of 93.75 %, 100 %, and 100 %, respectively, and specificities of 43.75 %, 62.5 %, and 18.75 %. The model designed to classify partial vs. full-thickness tears achieved an accuracy of 34.38 %. The model incorporating all pathological images compared to normal MRI images exhibited an accuracy of 37.50 % and a weighted F1 score of 0.32.</div></div><div><h3>Conclusion</h3><div>The results of the study suggest that, although CFDL applications may be promising for the initial detection of supraspinatus pathologies, their current iteration has limitations that must be resolved before they can be reliably integrated into clinical practice.</div></div>","PeriodicalId":44283,"journal":{"name":"Asia-Pacific Journal of Sport Medicine Arthroscopy Rehabilitation and Technology","volume":"42 ","pages":"Pages 1-7"},"PeriodicalIF":1.5000,"publicationDate":"2025-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Asia-Pacific Journal of Sport Medicine Arthroscopy Rehabilitation and Technology","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S221468732500010X","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ORTHOPEDICS","Score":null,"Total":0}
引用次数: 0

Abstract

Objective

To evaluate the performance of a code free deep learning (CFDL) application in diagnosing supraspinatus tendon pathologies on shoulder magnetic resonance imaging (MRI) images.

Design

This retrospective cross-sectional study included patients with supraspinatus MRI showing partial or full-thickness tears and tendinosis, with patients having normal findings as the control group. MRI images were processed in the LobeAI application using transfer learning with ResNet-50 V2 for model development. Models were built to differentiate each pathology from normal and full-thickness tears from partial tears.

Results

The ML models developed using the LobeAI application demonstrated the ability to differentiate between normal shoulder MRI images and partial tears, full-thickness tears, and tendinosis with sensitivities of 93.75 %, 100 %, and 100 %, respectively, and specificities of 43.75 %, 62.5 %, and 18.75 %. The model designed to classify partial vs. full-thickness tears achieved an accuracy of 34.38 %. The model incorporating all pathological images compared to normal MRI images exhibited an accuracy of 37.50 % and a weighted F1 score of 0.32.

Conclusion

The results of the study suggest that, although CFDL applications may be promising for the initial detection of supraspinatus pathologies, their current iteration has limitations that must be resolved before they can be reliably integrated into clinical practice.
使用无代码深度学习应用程序检测和分类肩部磁共振图像上的冈上肌病变
目的评价无代码深度学习(CFDL)在肩核磁共振成像(MRI)图像上诊断冈上肌腱病变的性能。设计:本回顾性横断面研究纳入冈上肌MRI显示部分或全层撕裂和肌腱萎缩的患者,对照组为表现正常的患者。在LobeAI应用程序中使用迁移学习与ResNet-50 V2进行模型开发,对MRI图像进行处理。建立模型以区分每种病理与正常撕裂和全层撕裂与部分撕裂。结果使用LobeAI应用程序建立的ML模型能够区分正常肩关节MRI图像与部分撕裂、全层撕裂和肌腱病,敏感性分别为93.75%、100%和100%,特异性分别为43.75%、62.5%和18.75%。该模型用于对部分和全层泪液进行分类,准确率为34.38%。与正常MRI图像相比,纳入所有病理图像的模型的准确率为37.50%,加权F1评分为0.32。结论本研究的结果表明,尽管CFDL应用于冈上肌病变的初步检测可能是有希望的,但其目前的迭代存在局限性,必须在可靠地整合到临床实践之前加以解决。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
CiteScore
3.80
自引率
0.00%
发文量
21
审稿时长
98 days
期刊介绍: The Asia-Pacific Journal of Sports Medicine, Arthroscopy, Rehabilitation and Technology (AP-SMART) is the official peer-reviewed, open access journal of the Asia-Pacific Knee, Arthroscopy and Sports Medicine Society (APKASS) and the Japanese Orthopaedic Society of Knee, Arthroscopy and Sports Medicine (JOSKAS). It is published quarterly, in January, April, July and October, by Elsevier. The mission of AP-SMART is to inspire clinicians, practitioners, scientists and engineers to work towards a common goal to improve quality of life in the international community. The Journal publishes original research, reviews, editorials, perspectives, and letters to the Editor. Multidisciplinary research with collaboration amongst clinicians and scientists from different disciplines will be the trend in the coming decades. AP-SMART provides a platform for the exchange of new clinical and scientific information in the most precise and expeditious way to achieve timely dissemination of information and cross-fertilization of ideas.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信