Artificial intelligence can be used in the identification and classification of shoulder osteoarthritis and avascular necrosis on plain radiographs: a training study of 7,139 radiograph sets
DOI:
https://doi.org/10.2340/17453674.2024.40905Keywords:
Avascular necrosis, Deep learning, Diagnostic accuracy, Radiological imaging, ShoulderAbstract
Background and purpose: Knowledge concerning the use AI models for the classification of glenohumeral osteoarthritis (GHOA) and avascular necrosis (AVN) of the humeral head is lacking. We aimed to analyze how a deep learning (DL) model trained to identify and grade GHOA on plain radiographs performs. Our secondary aim was to train a DL model to identify and grade AVN on plain radiographs.
Patients and methods: A modified ResNet-type network was trained on a dataset of radiographic shoulder examinations from a large tertiary hospital. A total of 7,139 radiographs were included. The dataset included various projections of the shoulder, and the network was trained using stochastic gradient descent. Performance evaluation metrics, area under the receiver operating characteristic curve (AUC), sensitivity, and specificity were used to assess the network’s performance for each outcome.
Results: The network demonstrated AUC values ranging from 0.73 to 0.93 for GHOA classification and > 0.90 for all AVN classification classes. The network exhibited lower AUC for mild cases compared with definitive cases of GHOA. When none and mild grades were combined, the AUC increased, suggesting difficulties in distinguishing between these 2 grades.
Conclusion: We found that a DL model can be trained to identify and grade GHOA on plain radiographs. Furthermore, we show that a DL model can identify and grade AVN on plain radiographs. The network performed well, particularly for definitive cases of GHOA and any level of AVN. However, challenges remain in distinguishing between none and mild GHOA grades.
Downloads
References
Willick S E, Sanders R K. Radiologic evaluation of the shoulder girdle. Phys Med Rehabil Clin N Am 2004; 15: 373-406. doi: 10.1016/j.pmr.2004.01.004. DOI: https://doi.org/10.1016/j.pmr.2004.01.004
Chee C G, Kim Y, Kang Y, Lee K J, Chae H-D, Cho J, et al. Performance of a deep learning algorithm in detecting osteonecrosis of the femoral head on digital radiography: a comparison with assessments by radiologists. AJR Am J Roentgenol 2019; 213: 155-62. doi: 10.2214/AJR.18.20817. DOI: https://doi.org/10.2214/AJR.18.20817
Bennell K, Hinman R. Exercise as a treatment for osteoarthritis. Curr Opin Rheumatol 2005; 17: 634-40. doi: 10.1097/01.bor.0000171214.49876.38. DOI: https://doi.org/10.1097/01.bor.0000171214.49876.38
Moretti L, Maccagnano G, Coviello M, Cassano G D, Franchini A, Laneve A, et al. Platelet rich plasma injections for knee osteoarthritis treatment: a prospective clinical study. J Clin Med 2022; 11: 2640. doi: 10.3390/jcm11092640. DOI: https://doi.org/10.3390/jcm11092640
Franceschi F, Franceschetti E, Paciotti M, Torre G, Samuelsson K, Papalia R, et al. Surgical management of osteonecrosis of the humeral head: a systematic review. Knee Surg Sports Traumatol Arthrosc 2017; 25: 3270-8. doi: 10.1007/s00167-016-4169-z. DOI: https://doi.org/10.1007/s00167-016-4169-z
Matthews A H, Davis D D, Fish M J, Stitson D. Avascular necrosis. StatPearls. Treasure Island (FL): StatPearls Publishing; 2023. Available from: http://www.ncbi.nlm.nih.gov/books/NBK537007/.
LeCun Y, Bengio Y, Hinton G. Deep learning. Nature 2015; 521: 436-44. doi: 10.1038/nature14539. DOI: https://doi.org/10.1038/nature14539
Sánchez-Rosenberg G, Magnéli M, Barle N, Kontakis M G, Müller A M, Wittauer M, et al. ChatGPT-4 generates orthopedic discharge documents faster than humans maintaining comparable quality: a pilot study of 6 cases. Acta Orthop 2024; 95: 152-6. doi: 10.2340/17453674.2024.40182. DOI: https://doi.org/10.2340/17453674.2024.40182
Norman B D, Pedoia V, Noworolski A, Link T M, Majumdar S. Automatic knee Kellgren Lawrence grading with artificial intelligence. Osteoarthritis Cartilage 2018; 26: S436-7. doi: 10.1016/j.joca.2018.02.840. DOI: https://doi.org/10.1016/j.joca.2018.02.840
Tiulpin A, Klein S, Bierma-Zeinstra S, Thevenot J, Meurs J van, Oei E, et al. Deep learning predicts knee osteoarthritis progression from plain radiographs. Osteoarthritis Cartilage 2019; 27: S397-8. doi: 10.1016/j.joca.2019.02.398. DOI: https://doi.org/10.1016/j.joca.2019.02.398
Kim D H, Lee K J, Choi D, Lee J I, Choi H G, Lee Y S. Can additional patient information improve the diagnostic performance of deep learning for the interpretation of knee osteoarthritis severity. J Clin Med 2020; 9: 3341. doi: 10.3390/jcm9103341. DOI: https://doi.org/10.3390/jcm9103341
von Schacky C E, Sohn J H, Liu F, Ozhinsky E, Jungmann P M, Nardo L, et al. Development and validation of a multitask deep learning model for severity grading of hip osteoarthritis features on radiographs. Radiology 2020; 295: 136-45. doi: 10.1148/radiol.2020190925.
Mongan J, Moy L, Kahn C E. Checklist for Artificial Intelligence in Medical Imaging (CLAIM): a guide for authors and reviewers. Radiol Artif Intell 2020; 2: e200029. doi: 10.1148/ryai.2020200029. DOI: https://doi.org/10.1148/ryai.2020200029
Macías-Hernández S I, Morones-Alba J D, Miranda-Duarte A, Coronado-Zarco R, Soria-Bastida M de L A, Nava-Bringas T, et al. Glenohumeral osteoarthritis: overview, therapy, and rehabilitation. Disabil Rehabil 2017; 39: 1674-82. doi: 10.1080/09638288.2016.1207206. DOI: https://doi.org/10.1080/09638288.2016.1207206
Sankaye P, Ostlere S. Arthritis at the shoulder joint. Semin Musculoskelet Radiol 2015; 19: 307-18. doi: 10.1055/s-0035-1549324. DOI: https://doi.org/10.1055/s-0035-1549324
Samilson R L, Prieto V. Dislocation arthropathy of the shoulder. J Bone Joint Surg Am 1983; 65: 456-60. PMID: 6833319 DOI: https://doi.org/10.2106/00004623-198365040-00005
Brox J I, Lereim P, Merckoll E, Finnanger A M. Radiographic classification of glenohumeral arthrosis. Acta Orthop Scand 2003; 74: 186-9. doi: 10.1080/00016470310013932. DOI: https://doi.org/10.1080/00016470310013932
Elsharkawi M, Cakir B, Reichel H, Kappe T. Reliability of radiologic glenohumeral osteoarthritis classifications. J Shoulder Elbow Surg 2013; 22: 1063-7. doi: 10.1016/j.jse.2012.11.007. DOI: https://doi.org/10.1016/j.jse.2012.11.007
Allain J, Goutallier D, Glorion C. Long-term results of the Latarjet procedure for the treatment of anterior instability of the shoulder. J Bone Joint Surg Am 1998; 80: 841-52. doi: 10.2106/00004623-199806000-00008. DOI: https://doi.org/10.2106/00004623-199806000-00008
Cruess R L. Steroid-induced avascular necrosis of the head of the humerus: natural history and management. J Bone Joint Surg Br. 1976; 58: 313-17. doi: 10.1302/0301-620X.58B3.956247. DOI: https://doi.org/10.1302/0301-620X.58B3.956247
Harreld K L, Marker D R, Wiesler E R, Shafiq B, Mont M A. Osteonecrosis of the humeral head. J Am Acad Orthop Surg 2009; 17: 345-55. doi: 10.5435/00124635-200906000-00003. DOI: https://doi.org/10.5435/00124635-200906000-00003
He K, Zhang X, Ren S, Sun J. Delving deep into rectifiers: surpassing human-level performance on ImageNet classification 2015. pp. 1026-34. Available from: https://www.cv-foundation.org/openaccess/content_iccv_2015/html/He_Delving_Deep_into_ICCV_2015_paper.html. DOI: https://doi.org/10.1109/ICCV.2015.123
Hajian-Tilaki K. Receiver operating characteristic (ROC) curve analysis for medical diagnostic test evaluation. Caspian J Intern Med 2013; 4: 627-35. PMID: 24009950
Youden W J. Index for rating diagnostic tests. Cancer 1950; 3: 32-5. doi: 10.1002/1097-0142(1950)3:1<32::aid-cncr2820030106>3.0.co;2-3. DOI: https://doi.org/10.1002/1097-0142(1950)3:1<32::AID-CNCR2820030106>3.0.CO;2-3
Grauhan N F, Niehues S M, Gaudin R A, Keller S, Vahldiek J L, Adams L C, et al. Deep learning for accurately recognizing common causes of shoulder pain on radiographs. Skeletal Radiol 2022; 51: 355-62. doi: 10.1007/s00256-021-03740-9. DOI: https://doi.org/10.1007/s00256-021-03740-9
von Schacky C E, Sohn J H, Liu F, Ozhinsky E, Jungmann P M, Nardo L, et al. Development and validation of a multitask deep learning model for severity grading of hip osteoarthritis features on radiographs. Radiology 2020; 295: 136-45. doi: 10.1148/radiol.2020190925. DOI: https://doi.org/10.1148/radiol.2020190925
Additional Files
Published
How to Cite
License
Copyright (c) 2024 Martin Magnéli, Michael Axenhus, Johan Fagrell, Petter Ling, Jacob Gislén, Yilmaz Demir, Erica Domeij-Arverud, Kristofer Hallberg, Björn Salomonsson, Max Gordon
This work is licensed under a Creative Commons Attribution 4.0 International License.
