Skip to main content

Advertisement

SpringerLink
Log in

Deep learning method for segmentation of rotator cuff muscles on MR images

  • Scientific Article
  • Published:
Skeletal Radiology Aims and scope Submit manuscript

Abstract

Objective

To develop and validate a deep convolutional neural network (CNN) method capable of (1) selecting a specific shoulder sagittal MR image (Y-view) and (2) automatically segmenting rotator cuff (RC) muscles on a Y-view. We hypothesized a CNN approach can accurately perform both tasks compared with manual reference standards.

Material and methods

We created 2 models: model A for Y-view selection and model B for muscle segmentation. For model A, we manually selected shoulder sagittal T1 Y-views from 258 cases as ground truth to train a classification CNN (Keras/Tensorflow, Inception v3, 16 batch, 100 epochs, dropout 0.2, learning rate 0.001, RMSprop). A top-3 success rate evaluated model A on 100 internal and 50 external test cases. For model B, we manually segmented subscapularis, supraspinatus, and infraspinatus/teres minor on 1048 sagittal T1 Y-views. After histogram equalization and data augmentation, the model was trained from scratch (U-Net, 8 batch, 50 epochs, dropout 0.25, learning rate 0.0001, softmax). Dice (F1) score determined segmentation accuracy on 105 internal and 50 external test images.

Results

Model A showed top-3 accuracy > 98% to select an appropriate Y-view. Model B produced accurate RC muscle segmentations with mean Dice scores > 0.93. Individual muscle Dice scores on internal/external datasets were as follows: subscapularis 0.96/0.93, supraspinatus 0.97/0.96, and infraspinatus/teres minor 0.97/0.95.

Conclusions

Our results show overall accurate Y-view selection and automated RC muscle segmentation using a combination of deep CNN algorithms.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6

Similar content being viewed by others

References

  1. Goutallier D, Postel J-M, Bernageau J, Lavau L, Voisin M-C. Fatty Muscle Degeneration in cuff ruptures: pre- and postoperative evaluation by CT scan. Clin Orthop. 1994;304:78–83.

    Google Scholar 

  2. Fuchs B, Weishaupt D, Zanetti M, Hodler J, Gerber C. Fatty degeneration of the muscles of the rotator cuff: assessment by computed tomography versus magnetic resonance imaging. J Shoulder Elb Surg. 1999;8(6):599–605.

    Article  CAS  Google Scholar 

  3. Mellado JM, Calmet J, Olona M, Esteve C, Camins A, Pérez Del Palomar L, et al. Surgically repaired massive rotator cuff tears: MRI of tendon integrity, muscle fatty degeneration, and muscle atrophy correlated with intraoperative and clinical findings. AJR Am J Roentgenol. 2005;184(5):1456–63.

    Article  CAS  Google Scholar 

  4. Jung W, Lee S, Hoon KS. The natural course of and risk factors for tear progression in conservatively treated full-thickness rotator cuff tears. J Shoulder Elb Surg. 2020;29(6):1168–1176.

  5. Wieser K, Joshy J, Filli L, Kriechling P, Sutter R, Fürnstahl P, et al. Changes of supraspinatus muscle volume and fat fraction after successful or failed arthroscopic rotator cuff repair. Am J Sports Med. 2019;47(13):3080–8.

    Article  Google Scholar 

  6. Kijowski R, Thurlow P, Blankenbaker D, Liu F, McGuine T, Li G, et al. Preoperative MRI shoulder findings associated with clinical outcome 1 year after rotator cuff repair. Radiology. 2019;291(3):722–9.

    Article  Google Scholar 

  7. Iannotti JP, Zlatkin MB, Esterhai JL, Kressel HY, Dalinka MK, Spindler KP. Magnetic resonance imaging of the shoulder. Sensitivity, specificity, and predictive value. J Bone Joint Surg Am. 1991;73(1):17–29.

    Article  CAS  Google Scholar 

  8. Lee SC, Williams D, Endo Y. The repaired rotator cuff: MRI and ultrasound evaluation. Curr Rev Musculoskelet Med. 2018;11(1):92–101.

    Article  Google Scholar 

  9. Zanetti M, Gerber C, Hodler J. Quantitative assessment of the muscles of the rotator cuff with magnetic resonance imaging. Investig Radiol. 1998;33(3):163–70.

    Article  CAS  Google Scholar 

  10. Thomazeau H, Rolland Y, Lucas C, Duval J-M, Langlais F. Atrophy of the supraspinatus belly. Assessment by MRI in 55 patients with rotator cuff pathology. Acta Orthop Scand. 1996;67(3):264–8.

    Article  CAS  Google Scholar 

  11. Di Benedetto P, Beltrame A, Cicuto C, Battistella C, Gisonni R, Cainero V, et al. Rotator cuff tears reparability index based on pre-operative MRI: our experience. Acta Bio Medica Atenei Parm. 2019;90(1-S):36–46.

    Google Scholar 

  12. Lehtinen J, Tingart M, Apreleva M, Zurakowski D, Palmer W, Warner J. Practical assessment of rotator cuff muscle volumes using shoulder MRI. Acta Orthop Scand. 2003;74(6):722–9.

    Article  Google Scholar 

  13. Lee Y-B, Yang C-J, Li CZ, Zhuan Z, Kwon S-C, Noh K-C. Can a single sagittal magnetic resonance imaging slice represent whole fatty infiltration in chronic rotator cuff tears at the supraspinatus? Clin Orthop Surg. 2018;10(1):55.

    Article  Google Scholar 

  14. Davis DL, Kesler T, Gilotra MN, Almardawi R, Hasan SA, Gullapalli RP, et al. Quantification of shoulder muscle intramuscular fatty infiltration on T1-weighted MRI: a viable alternative to the Goutallier classification system. Skelet Radiol. 2019;48(4):535–41.

    Article  Google Scholar 

  15. Matsumura N, Oguro S, Okuda S, Jinzaki M, Matsumoto M, Nakamura M, et al. Quantitative assessment of fatty infiltration and muscle volume of the rotator cuff muscles using 3-dimensional 2-point Dixon magnetic resonance imaging. J Shoulder Elb Surg. 2017;26(10):e309–18.

    Article  Google Scholar 

  16. Kim JY, Ro K, You S, Nam BR, Yook S, Park HS, et al. Development of an automatic muscle atrophy measuring algorithm to calculate the ratio of supraspinatus in supraspinous fossa using deep learning. Comput Methods Prog Biomed. 2019;182:105063.

    Article  Google Scholar 

  17. Conze P-H, Brochard S, Burdin V, Sheehan FT, Pons C. Healthy versus pathological learning transferability in shoulder muscle MRI segmentation using deep convolutional encoder-decoders. ArXiv190101620 Cs [Internet]. 2019 Feb 27 [cited 2020 Mar 23]; Available from: http://arxiv.org/abs/1901.01620.

  18. Wang Y-W, Lee C-C, Lo C-M. Supraspinatus segmentation from shoulder ultrasound images using a multilayer self-shrinking snake. IEEE Access. 2019;7:146724–31.

    Article  Google Scholar 

  19. Sezer A, Sezer HB. Capsule network-based classification of rotator cuff pathologies from MRI. Comput Electr Eng. 2019;80:106480.

    Article  Google Scholar 

  20. Szegedy C, Vanhoucke V, Ioffe S, Shlens J, Wojna Z. Rethinking the Inception architecture for computer vision. ArXiv151200567 Cs [Internet]. 2015 Dec 11 [cited 2020 Mar 31]; Available from: http://arxiv.org/abs/1512.00567.

  21. Ronneberger O, Fischer P, Brox T. U-Net: Convolutional networks for biomedical image segmentation. ArXiv150504597 Cs [Internet]. 2015 May 18 [cited 2020 Mar 23]; Available from: http://arxiv.org/abs/1505.04597.

  22. Abadi M, Barham P, Chen J, Chen Z, Davis A, Dean J, et al. TensorFlow: a system for large-scale machine learning. ArXiv160508695 Cs [Internet]. 2016 May 31 [cited 2020 Mar 23]; Available from: https://arxiv.org/abs/1605.08695.

  23. Dice LR. Measures of the amount of ecologic association between species. Ecology. 1945;26(3):297–302.

    Article  Google Scholar 

  24. Oh JH, Kim SH, Choi J-A, Kim Y, Oh CH. Reliability of the grading system for fatty degeneration of rotator cuff muscles. Clin Orthop. 2010;468(6):1558–64.

    Article  Google Scholar 

  25. Slabaugh MA, Friel NA, Karas V, Romeo AA, Verma NN, Cole BJ. Interobserver and intraobserver reliability of the Goutallier classification using magnetic resonance imaging: proposal of a simplified classification system to increase reliability. Am J Sports Med. 2012;40(8):1728–34.

    Article  Google Scholar 

  26. Wall LB, Teefey SA, Middleton WD, Dahiya N, Steger-May K, Kim HM, et al. Diagnostic performance and reliability of ultrasonography for fatty degeneration of the rotator cuff muscles. J Bone Joint Surg Am. 2012;94(12):e83.

    Article  Google Scholar 

  27. Schiefer M, Mendonça R, Magnanini MM, Fontenelle C, Pires Carvalho AC, Almeida M, et al. Intraobserver and interobserver agreement of Goutallier classification applied to magnetic resonance images. J Shoulder Elb Surg. 2015;24(8):1314–21.

    Article  Google Scholar 

  28. Chang S-C, Lee Y-W, Lai Y-C, Tiu C-M, Wang H-K, Chiou H-J, et al. Automatic slice selection and diagnosis of breast strain elastography. Med Phys. 2014;41(10):102902.

    Article  Google Scholar 

  29. Paknezhad M, Marchesseau S, Brown MS. Automatic basal slice detection for cardiac analysis. J Med Imaging [Internet]. 2016 Jul [cited 2020 Mar 31];3(3). Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5028419/

  30. Kanavati F, Islam S, Aboagye EO, Rockall A. Automatic L3 slice detection in 3D CT images using fully-convolutional networks. ArXiv181109244 Cs [Internet]. 2018 Nov 22 [cited 2020 Apr 2]; Available from: http://arxiv.org/abs/1811.09244

  31. Zhou Z, Zhao G, Kijowski R, Liu F. Deep convolutional neural network for segmentation of knee joint anatomy. Magn Reson Med. 2018;80(6):2759–70.

    Article  Google Scholar 

Download references

Author information

Author notes

  1. Giovanna Medina and Colleen G. Buckless contributed equally to this work.

Authors and Affiliations

  1. Department of Orthopedics, Massachusetts General Hospital, Harvard Medical School, Boston, MA, 02114, USA

    Giovanna Medina & Luke S. Oh

  2. Division of Musculoskeletal Imaging and Intervention, Department of Radiology, Massachusetts General Hospital, Harvard Medical School, 55 Fruit Street – YAW 6048, Boston, MA, 02114, USA

    Colleen G. Buckless, Eamon Thomasson & Martin Torriani

Authors
  1. Giovanna Medina
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Colleen G. Buckless
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Eamon Thomasson
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Luke S. Oh
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Martin Torriani
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Martin Torriani.

Ethics declarations

Conflict of interest

The authors declare that they have no conflicts of interest.

Ethical approval

All procedures performed in studies involving human participants were carried out in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards.

Informed consent

Informed consent was waived for individual participants included in the study. The study was approved by the local institutional review board (IRB) and HIPAA compliant.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Medina, G., Buckless, C.G., Thomasson, E. et al. Deep learning method for segmentation of rotator cuff muscles on MR images. Skeletal Radiol 50, 683–692 (2021). https://doi.org/10.1007/s00256-020-03599-2

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00256-020-03599-2

Keywords

Search

Navigation