肩關節旋轉肌撕裂傷好發於年長族群，臨床診斷工具包括肩關節鏡、核磁共振、超音波，然而上述診斷工具有高醫療支出或等待時間長久的缺點。X光影像具快速取得且低成本的特點，但並非為肩關節旋轉肌撕裂傷診斷的黃金標準。本研究開發基於X光影像辨識旋轉肌撕裂傷的可視化深度學習類神經網路：XCuffNet，並比較XCuffNet與經典深度學習網路如AlexNet、GoogLeNet、ResNet-50於識別旋轉肌撕裂傷X光影像的表現。本研究使用102筆肩關節的X光影像資料 (其中59筆資料確診為旋轉肌撕裂傷、43筆資料無旋轉肌撕裂傷)，XCuffNet辨識旋轉肌撕裂傷的準確率可達94.1%、效果較AlexNet、GoogLeNet、ResNet-50更佳。XCuffNet同時具備類別活化映射的可視化特色，XCuffNet經過臨床資料學習後、以肩峰(acromion)、大結節(greater tuberosity)、肩臼(glenoid)、肩峰與肱骨的區域(area between acromion and humeral)作為識別X光影像是否有旋轉肌撕裂傷的重要特徵。本研究同時將醫師判定之肩峰下骨刺型態(acromial spur type)、肩峰下骨刺尺寸(acromial spur size)、肩盂肱關節炎階段(glenohumeral joint arthritis stage)、肩峰間隔 (acromiohumeral interval)、Hamada類別(Hamada classification)，大結節型態(greater tuberosity morphology)、大結節骨刺厚度(GT spur thickness)組合為特徵向量，搭配機械學習演算法決策樹、支持向量機、K-近鄰演算法、集成學習演算法辨識X光影像資料是否為旋轉肌撕裂傷，達到92.2%的準確率。本研究開發的XCuffNet結合X光影像可快速且準確識別肩關節旋轉肌撕裂傷。

Shoulder plain radiography is a potential valuable tool to diagnose rotator cuff tears (RCTs) based on the radiographic parameters. Several radiographic parameters have been proposed as risks factors for atraumatic RCTs. The previous model for classifying RCTs from radiographs had good sensitivity but low specificity. This study was aimed toward developing a deep convolutional neural network (CNN) XCuffNet for RCT radiography classification, where machine learning and several representative CNN architectures were used to compare and evaluate the RCT radiography classification performance. Shoulder radiographs from 102 patients (59 RCT and 43 non-RCT) who underwent shoulder arthroscopy surgeries were retrospectively included. Seven radiographic parameters, including acromial spur type, acromial spur size, glenohumeral joint arthritis stage (Samilson Prieto (SPO) classification), acromiohumeral interval (AHI), rotator cuff arthropathy stage (Hamada classification), greater tuberosity (GT) morphology, and GT spur thickness, were selected using sequential feature selection for a machine learning algorithm. Decision tree, support vector machine (SVM), k-nearest neighbor (KNN), and ensemble learning methods were used in this study. A deep CNN, XCuffNet, was developed in this study and compared with state-of-the-art CNNs, including AlexNet, GoogLeNet, and ResNet-50, which were trained based on shoulder radiographs. The accuracy, sensitivity, specificity, precision, F-1 score, and area under the receiver operating characteristic curves (AUC) of the machine learning and deep learning algorithm were measured. The deep CNN XCuffNet and several CNNs outperformed machine learning in classifying RCT and obtained 92.2% and 94.1% accuracy in both machine learning and deep learning. However, good accuracy, high sensitivity, high specificity, high precision, and high F-1 scores were obtained using both deep learning and machine learning algorithms for classifying RCTs based on shoulder radiographs and the radiographic features. Acromial spur type, AHI, Hamada classification, and GT morphology were selected as representative features of RCT radiography classification using sequential forward selection (SFS), sequential backward selection (SBS), and neighborhood component analysis (NCA) feature selection. The proposed XCuffNet method improved the existing CNN performance and obtained an accuracy of 94.1%. The representative areas for classifying RCT radiography include the acromion, greater tuberosity, glenoid, and area between the acromion and humeral based on feature visualization that correlates with the shoulder radiographic features of RCTs. Acromial spur type, SPO classification, AHI, Hamada classification, and GT morphology were computed automatically to classify RCTs in the future.

[1] S. N. Sambandam, V. Khanna, A. Gul, and V. Mounasamy, “Rotator cuff tears: An evidence based approach,” World journal of orthopedics, vol. 6, no. 11, p. 902, 2015.
[2] A. Yamamoto, K. Takagishi, T. Osawa, T. Yanagawa, D. Nakajima, H. Shitara, and T. Kobayashi, “Prevalence and risk factors of a rotator cuff tear in the general population,” Journal of shoulder and elbow surgery, vol. 19, no. 1, pp. 116-120, 2010.
[3] P.-C. Nolte, M. Seiter, B. P. Elrick, and P. J. Millett, “Rotator cuff tears in elderly patients: is repair worthwhile?,” 2020.
[4] R. Cofield, “Subscapular muscle transposition for repair of chronic rotator cuff tears,” Surgery, gynecology & obstetrics, vol. 154, no. 5, pp. 667-672, 1982.
[5] A. Lädermann, P. J. Denard, and P. Collin, “Massive rotator cuff tears: definition and treatment,” International orthopaedics, vol. 39, no. 12, pp. 2403-2414, 2015.
[6] A. K. Sørensen, K. Bak, A. L. Krarup, C. H. Thune, M.Nygaard, U. Jørgensen, C. Sloth, and S. T.-Pedersen, “Acute rotator cuff tear: do we miss the early diagnosis? A prospective study showing a high incidence of rotator cuff tears after shoulder trauma,” Journal of shoulder and elbow surgery, vol. 16, no. 2, pp. 174-180, 2007.
[7] S. Weber and J. Chahal, “Case Studies AAOS Clinical Practice Guideline: Management of Rotator Cuff Injuries,” JAAOS-Journal of the American Academy of Orthopaedic Surgeons, p. 10.5435, 2021.
[8] S. Frangiamore, G. J. Dornan, M. P. Horan, S. Mannava, E. M. Fritz, Z. B. Hussain, G. Moatshe, J. A. Godin, J. Pogorzelski, and P. J. Millett, “Predictive Modeling to Determine Functional Outcomes After Arthroscopic Rotator Cuff Repair,” The American journal of sports medicine, vol. 48, no. 7, pp. 1559-1567, 2020.
[9] F. Plachel, P. Siegert, K. Rüttershoff, K. Thiele, D. Akgün, P. Moroder, M. Scheibel, C. Gerhardt, “Long-term results of arthroscopic rotator cuff repair: A follow-up study comparing single-row versus double-row fixation techniques,” The American journal of sports medicine, vol. 48, no. 7, pp. 1568-1574, 2020.
[10] S. Ø. Gismervik, J. O. Drogset, F. Granviken, M. Rø, and G. Leivseth, “Physical examination tests of the shoulder: a systematic review and meta-analysis of diagnostic test performance,” BMC musculoskeletal disorders, vol. 18, no. 1, pp. 1-9, 2017.
[11] M. H. Lee, S. E. Sheehan, J. F. Orwin, and K. S. Lee, “Comprehensive shoulder US examination: a standardized approach with multimodality correlation for common shoulder disease,” Radiographics, vol. 36, no. 6, pp. 1606-1627, 2016.
[12] A. P. Apostolopoulos, S. Angelis, R. K. Yallapragada, S. Khan, J. Nadjafi, T. Balfousias, and T. P. Selvan, “The Sensitivity of Magnetic Resonance Imaging and Ultrasonography in Detecting Rotator Cuff Tears,” Cureus, vol. 11, no. 5, 2019.
[13] H. Hapani, M. Sood, A. Trivedi, A. Chawla, I. Virda, and K. Radadiya, “Ultrasound, MRI and Arthroscopic Correlation of Rotator Cuff Tears,” International Journal of Contemporary Medical Research, vol. 4, no. 3, pp. 77-83, 2017.
[14] K. R. Okoroha, N. Mehran, J. Duncan, T. Washington, T. Spiering, M. J. Bey, M. V. Holsbeeck, and V. Moutzouros, “Characterization of rotator cuff tears: ultrasound versus magnetic resonance imaging,” Orthopedics, vol. 40, no. 1, pp. e124-e130, 2017.
[15] H.-Y. Lu, C.-Y. Huang, C.-T. Su, and C.-C. Lin, “Predicting rotator cuff tears using data mining and bayesian likelihood ratios,” PloS one, vol. 9, no. 4, p. e94917, 2014.
[16] H.-C. Chuang, C.-K. Hong, K.-L. Hsu, F.-C. Kuan, C.-L. Lin, and W.-R. Su, “The radiographic morphology of the greater tuberosity is associated with muscle degeneration in patients with symptomatic rotator cuff tears,” Journal of shoulder and elbow surgery, vol. 28, no. 10, pp. 1964-1970, 2019.
[17] K. Ogawa, A. Yoshida, W. Inokuchi, and T. Naniwa, “Acromial spur: relationship to aging and morphologic changes in the rotator cuff,” Journal of shoulder and elbow surgery, vol. 14, no. 6, pp. 591-598, 2005.
[18] H. B. Park, J.-Y. Gwark, B. H. Kwack, and J.-B. Na, “Are Any Radiologic Parameters Independently Associated with Degenerative PSRCT-Significance of Critical Shoulder Angle,” Journal of Shoulder and Elbow Surgery, 2020.
[19] J. H. Oh, J. Y. Kim, H. K. Lee, and J.-A. Choi, “Classification and clinical significance of acromial spur in rotator cuff tear: heel-type spur and rotator cuff tear,” Clinical Orthopaedics and Related Research®, vol. 468, no. 6, pp. 1542-1550, 2010.
[20] T. M. Ghandour, S. A. Elghazaly, and A. M. Ghandour, “Greater tuberosity sclerosis: a radiographic sign of rotator cuff tear,” Acta Orthop Belg, vol. 83, pp. 416-420, 2015.
[21] H. Razmjou, V. Palinkas, M. Christakis, S. Robarts, and D. Kennedy, “Reduced acromiohumeral distance and increased critical shoulder angle: implications for primary care clinicians,” The Physician and sportsmedicine, vol. 48, no. 3, pp. 312-319, 2020.
[22] N. Saupe, C. W. Pfirrmann, M. R. Schmid, B. Jost, C. M. Werner, and M. Zanetti, “Association between rotator cuff abnormalities and reduced acromiohumeral distance,” American Journal of Roentgenology, vol. 187, no. 2, pp. 376-382, 2006.
[23] P. N. Chalmers, D. H. Salazar, K. S. May, A. M. Chamberlain, G. S.-Cucchi, K. Yamaguchi, and J. D. Keener, “Radiographic progression of arthritic changes in shoulders with degenerative rotator cuff tears,” Journal of shoulder and elbow surgery, vol. 25, no. 11, pp. 1749-1755, 2016.
[24] R. Samilson and V. Prieto, “Dislocation arthropathy of the shoulder,” The Journal of bone and joint surgery. American volume, vol. 65, no. 4, pp. 456-460, 1983.
[25] H.C. Chuang, C. K. Hong, K. L. Hsu, F. C. Kuan, C. H. Chiang, Y. Chen, and W. R. Su , “Radiographic greater tuberosity spurs and narrow acromiohumeral intervals are associated with advanced retraction of the supraspinatus tendon in patients with symptomatic rotator cuff tears,” JSES international, vol. 5, no. 1, pp. 77-82, 2021.
[26] T. J. Brolin, G. F. Updegrove, and J. G. Horneff, “Classifications in brief: Hamada classification of massive rotator cuff tears,” Clinical Orthopaedics and Related Research®, vol. 475, no. 11, pp. 2819-2823, 2017.
[27] F. Familiari, S. K. Nayar, R. Russo, M. D. Gori, F. Ranuccio, V. Mastroianni, E. Giuzio, O. Galasso, G. Gasparini, E. G. McFarland, U. Srikumaran, “Subacromial balloon spacer for massive, irreparable rotator cuff tears is associated with improved shoulder function and high patient satisfaction,” Arthroscopy: The Journal of Arthroscopic & Related Surgery, vol. 37, no. 2, pp. 480-486, 2021.
[28] S. S. Burkhart, J. J. Pranckun, and R. U. Hartzler, “Superior capsular reconstruction for the operatively irreparable rotator cuff tear: Clinical outcomes are maintained 2 years after surgery,” Arthroscopy: The Journal of Arthroscopic & Related Surgery, vol. 36, no. 2, pp. 373-380, 2020.
[29] F. C. Golding, “The shoulder—the forgotten joint. The Makenzie Davidson memorial lecture,” The British journal of radiology, vol. 35, no. 411, pp. 149-158, 1962.
[30] R. Mirzayan, S. Donohoe, M. Batech, B. D. Suh, D. C. Acevedo, and A. Singh, “Is there a difference in the acromiohumeral distances measured on radiographic and magnetic resonance images of the same shoulder with a massive rotator cuff tear?,” Journal of shoulder and elbow surgery, vol. 29, no. 6, pp. 1145-1151, 2020.
[31] G. Bernhardt, M. Glehr, M. Zacherl, C. Wurnig, and G. Gruber, “Observer variability in the assessment of the acromiohumeral interval using anteroposterior shoulder radiographs,” European Journal of Orthopaedic Surgery & Traumatology, vol. 23, no. 2, pp. 185-190, 2013.
[32] J. Bullock, C. Cuesta-Lázaro, and A. Quera-Bofarull, “XNet: A convolutional neural network (CNN) implementation for medical X-ray image segmentation suitable for small datasets,” in Medical Imaging 2019: Biomedical Applications in Molecular, Structural, and Functional Imaging, vol. 10953: International Society for Optics and Photonics, p. 109531Z, 2019.
[33] A. Pemasiri, K. Nguyen, S. Sridharan, and C. Fookes, “Multi-modal semantic image segmentation,” Computer Vision and Image Understanding, vol. 202, p. 103085, 2021.
[34] C.-C. Lin, C.-N. Wang, Y.-K. Ou, and J. Fu, “Combined image enhancement, feature extraction, and classification protocol to improve detection and diagnosis of rotator-cuff tears on MR imaging,” Magnetic Resonance in Medical Sciences, vol. 13, no. 3, pp. 155-166, 2014.
[35] E. Shim, J. Y. Kim, J. P. Yoon, S. Y. Ki, T. Lho, Y. Kim, and S. W. Chung, “Automated rotator cuff tear classification using 3D convolutional neural network,” Scientific Reports, vol. 10, no. 1, pp. 1-9, 2020.
[36] Y. Kim., D. Choi, K. J. Lee, Y. Kang, J. M. Ahn, E. Lee, J. W. Lee, and H. S. Kang. “Ruling out rotator cuff tear in shoulder radiograph series using deep learning: redefining the role of conventional radiograph,” European radiology, pp. 1-10, 2020.
[37] A. Marcano-Cedeño, J. Quintanilla-Domínguez, M. Cortina-Januchs, and D. Andina, “Feature selection using sequential forward selection and classification applying artificial metaplasticity neural network,” in IECON 2010-36th annual conference on IEEE industrial electronics society: IEEE, pp. 2845-2850, 2010.
[38] L. Burrell, O. Smart, G. K. Georgoulas, E. D. Marsh, and G. J. Vachtsevanos, “Evaluation of Feature Selection Techniques for Analysis of Functional MRI and EEG,” in DMIN, pp. 256-262, 2007.
[39] W. Yang, K. Wang, and W. Zuo, “Neighborhood component feature selection for high-dimensional data,” JCP, vol. 7, no. 1, pp. 161-168, 2012.
[40] Q. D. N. Nguyen, A.-B. Liu, and C.-W. Lin, “Development of a Neurodegenerative Disease Gait Classification Algorithm Using Multiscale Sample Entropy and Machine Learning Classifiers,” Entropy, vol. 22, no. 12, p. 1340, 2020.
[41] J. Fairley, G. Georgoulas, and G. Vachtsevanos, “Sequential feature selection methods for Parkinsonian human sleep analysis,” in 2009 17th Mediterranean Conference on Control and Automation: IEEE, pp. 1468-1473, 2009.
[42] O. Yaman, “An automated faults classification method based on binary pattern and neighborhood component analysis using induction motor,” Measurement, vol. 168, p. 108323, 2021.
[43] W. S. Noble, “What is a support vector machine?,” Nature biotechnology, vol. 24, no. 12, pp. 1565-1567, 2006.
[44] A. Widodo and B.-S. Yang, “Support vector machine in machine condition monitoring and fault diagnosis,” Mechanical systems and signal processing, vol. 21, no. 6, pp. 2560-2574, 2007.
[45] Z. Zhang, “Introduction to machine learning: k-nearest neighbors,” Annals of translational medicine, vol. 4, no. 11, 2016.
[46] N. S. Altman, “An introduction to kernel and nearest-neighbor nonparametric regression,” The American Statistician, vol. 46, no. 3, pp. 175-185, 1992.
[47] V. Podgorelec, P. Kokol, B. Stiglic, and I. Rozman, “Decision trees: an overview and their use in medicine,” Journal of medical systems, vol. 26, no. 5, pp. 445-463, 2002.
[48] O. Yaman, H. Yetis, and M. Karakose, “Decision Tree Based Customer Analysis Method for Energy Planning in Smart Cities,” in 2020 International Conference on Data Analytics for Business and Industry: Way Towards a Sustainable Economy (ICDABI): IEEE, pp. 1-4, 2020.
[49] Y. Freund and L. Mason, “The alternating decision tree learning algorithm,” in icml, vol. 99: Citeseer, pp. 124-133, 1999.
[50] T. G. Dietterich, “Ensemble methods in machine learning,” in International workshop on multiple classifier systems: Springer, pp. 1-15, 2000.
[51] A. S. Ashour, Y. Guo, A. R. Hawas, and G. Xu, “Ensemble of subspace discriminant classifiers for schistosomal liver fibrosis staging in mice microscopic images,” Health information science and systems, vol. 6, no. 1, pp. 1-10, 2018.
[52] S. Sun, J. Pang, J. Shi, S. Yi, and W. Ouyang, “Fishnet: A versatile backbone for image, region, and pixel level prediction,” arXiv preprint arXiv:1901.03495, 2019.
[53] W. Rawat and Z. Wang, “Deep convolutional neural networks for image classification: A comprehensive review,” Neural computation, vol. 29, no. 9, pp. 2352-2449, 2017.
[54] A. Krizhevsky, I. Sutskever, and G. E. Hinton, “Imagenet classification with deep convolutional neural networks,” Advances in neural information processing systems, vol. 25, pp. 1097-1105, 2012.
[55] C. Szegedy, W. Liu, Y. Jia, P. Sermanet, S. Reed, D. Anguelov, D. Erhan, V. Vanhoucke, and A. Rabinovich, “Going deeper with convolutions,” in Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 1-9, 2015.
[56] K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” in Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 770-778, 2016.
[57] B. Zhou, A. Khosla, A. Lapedriza, A. Oliva, and A. Torralba, “Learning deep features for discriminative localization,” in Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 2921-2929, 2016.
[58] J. T. Springenberg, A. Dosovitskiy, T. Brox, and M. Riedmiller, “Striving for simplicity: The all convolutional net,” arXiv preprint arXiv:1412.6806, 2014.
[59] K. Simonyan, A. Vedaldi, and A. Zisserman, “Deep inside convolutional networks: Visualising image classification models and saliency maps,” arXiv preprint arXiv:1312.6034, 2013.
[60] M. D. Zeiler and R. Fergus, “Visualizing and understanding convolutional networks,” in European conference on computer vision: Springer, pp. 818-833, 2014.
[61] H. Sun, X. Zeng, T. Xu, G. Peng, and Y. Ma, “Computer-aided diagnosis in histopathological images of the endometrium using a convolutional neural network and attention mechanisms,” IEEE journal of biomedical and health informatics, vol. 24, no. 6, pp. 1664-1676, 2019.
[62] W. Hu, X. Meng, Y. Bai, A. Zhang, B. Cai, G. Zhang, T. W. Wilson, J. M. Stephen, V. D. Calhoun, and Y. P. Wang, “Interpretable multimodal fusion networks reveal mechanisms of brain cognition,” IEEE transactions on medical imaging, vol. 40, no. 5, pp. 1474-1483, 2021.
[63] C. Hu and Y. Wang, “An Efficient Convolutional Neural Network Model Based on Object-Level Attention Mechanism for Casting Defect Detection on Radiography Images,” IEEE Transactions on Industrial Electronics, vol. 67, no. 12, pp. 10922-10930, 2020.
[64] R. R. Selvaraju, M. Cogswell, A. Das, R. Vedantam, D. Parikh, and D. Batra, “Grad-cam: Visual explanations from deep networks via gradient-based localization,” in Proceedings of the IEEE international conference on computer vision, 2017, pp. 618-626, 2017.
[65] G. G. Godinho, F. M. L. Santos, F. O. França, J. M. A. Freitas, F. A. S. Mesquita, and T. S. d. A. Silva, “Videoarthroscopic treatment of glenohumeral osteoarthritis,” Revista brasileira de ortopedia, vol. 48, no. 1, pp. 69-79, 2013.
[66] G. G. Godinho, F. M. L. Santos, F. O. França, J. M. A. Freitas, F. A. S. Mesquita, and T. S. d. A. Silva, “Videoarthroscopic treatment of glenohumeral osteoarthritis,” Revista brasileira de ortopedia, vol. 48, pp. 69-79, 2013.
[67] A. Garcia-Garcia, S. Orts-Escolano, S. Oprea, V. Villena-Martinez, and J. Garcia-Rodriguez, “A review on deep learning techniques applied to semantic segmentation,” arXiv preprint arXiv:1704.06857, 2017.
[68] W. Yang, J. Zhang, Z. Xu, and K. Hu, “Real-time DeepLabv3+ for pedestrian segmentation,” Journal of Optical Technology, vol. 86, no. 9, pp. 570-578, 2019.
[69] L.-C. Chen, Y. Zhu, G. Papandreou, F. Schroff, and H. Adam, “Encoder-decoder with atrous separable convolution for semantic image segmentation,” in Proceedings of the European conference on computer vision (ECCV), pp. 801-818, 2018.
[70] A. G. Howard, M. Zhu, B. Chen, D. Kalenichenko, W. Wang, T. Weyand, M. Andreetto, and H. Adam, “Mobilenets: Efficient convolutional neural networks for mobile vision applications,” arXiv preprint arXiv:1704.04861, 2017.
[71] F. Gong, B. Gao, and Q. Niu, “An algorithm for rapidly computing the minimum distance between two objects collision detection,” in 2008 Congress on Image and Signal Processing, vol. 2: IEEE, pp. 676-679, 2008.
[72] C. Sammut and G. I. Webb, Encyclopedia of machine learning and data mining. Springer, 2017.
[73] S. Landau, A handbook of statistical analyses using SPSS. Chapman & Hall, 2019.
[74] D. Arkkelin, “Using SPSS to understand research and data analysis,” 2014.
[75] K. Hamada, K. Yamanaka, Y. Uchiyama, T. Mikasa, and M. Mikasa, “A radiographic classification of massive rotator cuff tear arthritis,” Clinical Orthopaedics and Related Research®, vol. 469, no. 9, pp. 2452-2460, 2011.
[76] H. Cheng, D. J. Garrick, and R. L. Fernando, “Efficient strategies for leave-one-out cross validation for genomic best linear unbiased prediction,” Journal of animal science and biotechnology, vol. 8, no. 1, pp. 1-5, 2017.
[77] W. Zhu, N. Zeng, and N. Wang, “Sensitivity, specificity, accuracy, associated confidence interval and ROC analysis with practical SAS implementations,” NESUG proceedings: health care and life sciences, Baltimore, Maryland, vol. 19, p. 67, 2010.
[78] T. Fawcett, “An introduction to ROC analysis,” Pattern recognition letters, vol. 27, no. 8, pp. 861-874, 2006.
[79] J. Hintze, “NCSS statistical software,” NCSS, Kaysville, UT, 1998.
[80] J. Y. Choi, J. K. Yum, and M. C. Song, “Correlation between degree of torn rotator cuff in MRI and degenerative change of acromion and greater tuberosity in simple radiography,” Clinics in Shoulder and Elbow, vol. 16, no. 1, pp. 1-9, 2013.
[81] G. Gruber, G. A. Bernhardt, H. Clar, M. Zacherl, M. Glehr, and C. Wurnig, “Measurement of the acromiohumeral interval on standardized anteroposterior radiographs: a prospective study of observer variability,” Journal of shoulder and elbow surgery, vol. 19, no. 1, pp. 10-13, 2010.
[82] S. Landau and B. S. Everitt, A handbook of statistical analyses using SPSS. Chapman and Hall/CRC, 2003.
[83] K. Shinagawa, T. Hatta, N. Yamamoto, J. Kawakami, Y. Shiota, M. Mineta, and E. Itoi, “Critical shoulder angle in an East Asian population: correlation to the incidence of rotator cuff tear and glenohumeral osteoarthritis,” Journal of shoulder and elbow surgery, vol. 27, no. 9, pp. 1602-1606, 2018.