Top

Published in:

Open Access 01-01-2021 | Computed Tomography | Musculoskeletal

Deep learning for the rapid automatic quantification and characterization of rotator cuff muscle degeneration from shoulder CT datasets

Authors: Elham Taghizadeh, Oskar Truffer, Fabio Becce, Sylvain Eminian, Stacey Gidoin, Alexandre Terrier, Alain Farron, Philippe Büchler

Published in: European Radiology | Issue 1/2021

Abstract

Objectives

This study aimed at developing a convolutional neural network (CNN) able to automatically quantify and characterize the level of degeneration of rotator cuff (RC) muscles from shoulder CT images including muscle atrophy and fatty infiltration.

Methods

One hundred three shoulder CT scans from 95 patients with primary glenohumeral osteoarthritis undergoing anatomical total shoulder arthroplasty were retrospectively retrieved. Three independent radiologists manually segmented the premorbid boundaries of all four RC muscles on standardized sagittal-oblique CT sections. This premorbid muscle segmentation was further automatically predicted using a CNN. Automatically predicted premorbid segmentations were then used to quantify the ratio of muscle atrophy, fatty infiltration, secondary bone formation, and overall muscle degeneration. These muscle parameters were compared with measures obtained manually by human raters.

Results

Average Dice similarity coefficients for muscle segmentations obtained automatically with the CNN (88% ± 9%) and manually by human raters (89% ± 6%) were comparable. No significant differences were observed for the subscapularis, supraspinatus, and teres minor muscles (p > 0.120), whereas Dice coefficients of the automatic segmentation were significantly higher for the infraspinatus (p < 0.012). The automatic approach was able to provide good–very good estimates of muscle atrophy (R² = 0.87), fatty infiltration (R² = 0.91), and overall muscle degeneration (R² = 0.91). However, CNN-derived segmentations showed a higher variability in quantifying secondary bone formation (R² = 0.61) than human raters (R² = 0.87).

Conclusions

Deep learning provides a rapid and reliable automatic quantification of RC muscle atrophy, fatty infiltration, and overall muscle degeneration directly from preoperative shoulder CT scans of osteoarthritic patients, with an accuracy comparable with that of human raters.

Key Points

• Deep learning can not only segment RC muscles currently available in CT images but also learn their pre-existing locations and shapes from invariant anatomical structures visible on CT sections.

• Our automatic method is able to provide a rapid and reliable quantification of RC muscle atrophy and fatty infiltration from conventional shoulder CT scans.

• The accuracy of our automatic quantitative technique is comparable with that of human raters.

Goutallier D, Postel JM, Bernageau J, Lavau L, Voisin MC (1994) Fatty muscle degeneration in cuff ruptures. Pre- and postoperative evaluation by CT scan. Clin Orthop Relat Res:78–83. https://doi.org/10.1097/00003086-199407000-00014

Lapner PLC, Jiang L, Zhang T, Athwal GS (2015) Rotator cuff fatty infiltration and atrophy are associated with functional outcomes in anatomic shoulder arthroplasty. Clin Orthop Relat Res 473:674–682. https://doi.org/10.1007/s11999-014-3963-5CrossRefPubMed

Donohue KW, Ricchetti ET, Ho JC, Iannotti JP (2018) The association between rotator cuff muscle fatty infiltration and glenoid morphology in glenohumeral osteoarthritis. J Bone Joint Surg Am 100:381–387. https://doi.org/10.2106/JBJS.17.00232CrossRefPubMed

McElvany MD, McGoldrick E, Gee AO, Neradilek MB, Matsen 3rd FA (2015) Rotator cuff repair: published evidence on factors associated with repair integrity and clinical outcome. Am J Sports Med 43:491–500. https://doi.org/10.1177/0363546514529644

Gladstone JN, Bishop JY, Lo IKY, Flatow EL (2007) Fatty infiltration and atrophy of the rotator cuff do not improve after rotator cuff repair and correlate with poor functional outcome. Am J Sports Med 35:719–728. https://doi.org/10.1177/0363546506297539CrossRefPubMed

van de Sande MAJ, Stoel BC, Obermann WR, Tjong a Lieng JGS, Rozing PM (2005) Quantitative assessment of fatty degeneration in rotator cuff muscles determined with computed tomography. Invest Radiol 40:313–319. https://doi.org/10.1097/01.rli.0000160014.16577.86

Terrier A, Ston J, Dewarrat A, Becce F, Farron A (2017) A semi-automated quantitative CT method for measuring rotator cuff muscle degeneration in shoulders with primary osteoarthritis. Orthop Traumatol Surg Res 103:151–157. https://doi.org/10.1016/j.otsr.2016.12.006

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

Lin DJ, Wong TT, Kazam JK (1994) Shoulder arthroplasty, from indications to complications: what the radiologist needs to know. Radiographics 36:192–208. https://doi.org/10.1148/rg.2016150055CrossRef

10.

Dekker TJ, Steele JR, Vinson EV, Garrigues GE (2019) Current peri-operative imaging concepts surrounding shoulder arthroplasty. Skeletal Radiol 48:1485–1497. https://doi.org/10.1007/s00256-019-03183-3CrossRefPubMed

11.

Buck FM, Jost B, Hodler J (2008) Shoulder arthroplasty. Eur Radiol 18:2937–2948. https://doi.org/10.1007/s00330-008-1093-8CrossRefPubMed

12.

Thomazeau H, Rolland Y, Lucas C, Duval JM, Langlais F (1996) Atrophy of the supraspinatus belly: assessment by MRI in 55 patients with rotator cuff pathology. Acta Orthop Scand 67:264–268. https://doi.org/10.3109/17453679608994685

13.

Zanetti M, Gerber C, Hodler J (1998) Quantitative assessment of the muscles of the rotator cuff with magnetic resonance imaging. Invest Radiol 33:163–170CrossRefPubMed

14.

Oh JH, Kim SH, Choi JA, Kim Y, Oh CH (2010) Reliability of the grading system for fatty degeneration of rotator cuff muscles. Clin Orthop Relat Res 468:1558–1564. https://doi.org/10.1007/s11999-009-0818-6

15.

Slabaugh MA, Friel NA, Karas V, Romeo AA, Verma NN, Brian J Cole BJ (2012) 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 40:1728–1734. https://doi.org/10.1177/0363546512452714

16.

Hashimoto F, Kakimoto A, Ota N, Ito S, Nishizawa S (2019) Automated segmentation of 2D low-dose CT images of the psoas-major muscle using deep convolutional neural networks. Radiol Phys Technol 12:210–215. https://doi.org/10.1007/s12194-019-00512-y

17.

Weber KA, Smith AC, Wasielewski M et al (2019) Deep learning convolutional neural networks for the automatic quantification of muscle fat infiltration following whiplash injury. Sci Rep 9:7973. https://doi.org/10.1038/s41598-019-44416-8CrossRefPubMedPubMedCentral

18.

Burns JE, Yao J, Chalhoub D, Chen JJ, Summers RM (2019) A machine learning algorithm to estimate sarcopenia on abdominal CT. Acad Radiol:1–10. https://doi.org/10.1016/j.acra.2019.03.011

19.

Graffy PM, Liu J, Pickhardt PJ, Burns JE, Yao J, Summers RM (2019) Deep learning-based muscle segmentation and quantification at abdominal CT: application to a longitudinal adult screening cohort for sarcopenia assessment. Br J Radiol:20190327. https://doi.org/10.1259/bjr.20190327

20.

Kim Y, Choi D, Lee KJ et al (2020) Ruling out rotator cuff tear in shoulder radiograph series using deep learning: redefining the role of conventional radiograph. Eur Radiol 30:2843–2852. https://doi.org/10.1007/s00330-019-06639-1CrossRefPubMed

21.

Terrier A, Ston J, Larrea X, Farron A (2014) Measurements of three-dimensional glenoid erosion when planning the prosthetic replacement of osteoarthritic shoulders. Bone Joint J 96-B:513–518. https://doi.org/10.1302/0301-620X.96B4.32641CrossRefPubMed

22.

Terrier A, Ston J, Farron A (2015) Importance of a three-dimensional measure of humeral head subluxation in osteoarthritic shoulders. J Shoulder Elbow Surg 24:295–301. https://doi.org/10.1016/j.jse.2014.05.027CrossRefPubMed

23.

Ronneberger O, Fischer P, Brox T (2015) U-net: convolutional networks for biomedical image segmentation. Lect Notes Comput Sci (including Subser Lect Notes Artif Intell Lect Notes Bioinformatics) 9351:234–241. https://doi.org/10.1007/978-3-319-24574-4_28CrossRef

24.

Ioffe S, Szegedy C (2015) Batch normalization: accelerating deep network training by reducing internal covariate shift. J Can Dent Assoc 70:156–157

25.

Warfield SK, Zou KH, Wells WM (2004) Simultaneous truth and performance level estimation (STAPLE): an algorithm for the validation of image segmentation. IEEE Trans Med Imaging 23:903–921. https://doi.org/10.1109/TMI.2004.828354CrossRefPubMedPubMedCentral

26.

Williams MD, Lädermann A, Melis B, Barthelemy R, Walch G (2009) Fatty infiltration of the supraspinatus: a reliability study. J Shoulder Elbow Surg 18:581–587. https://doi.org/10.1016/j.jse.2008.12.014

27.

Hemke R, Buckless CG, Tsao A, Wang B, Torriani M (2019) Deep learning for automated segmentation of pelvic muscles, fat, and bone from CT studies for body composition assessment. Skeletal Radiol. https://doi.org/10.1007/s00256-019-03289-8

28.

Kim S, Lee D, Park S, Oh K-S, Chung SW, Kim Y (2017) Automatic segmentation of supraspinatus from MRI by internal shape fitting and autocorrection. Comput Methods Programs Biomed 140:165–174. https://doi.org/10.1016/j.cmpb.2016.12.008

29.

Ascani D, Mazzà C, De Lollis A, Bernardoni M, Viceconti M (2015) A procedure to estimate the origins and the insertions of the knee ligaments from computed tomography images. J Biomech 48:233–237. https://doi.org/10.1016/j.jbiomech.2014.11.041

30.

de Oliveira ME, Netto LMG, Kistler M, Brandenberger D, Büchler P, Hasler C-C (2014) An image-based method to automatically propagate bony landmarks: application to computational spine biomechanics. Comput Methods Biomech Biomed Eng:1–8. https://doi.org/10.1080/10255842.2014.927445

31.

Taghizadeh E, Terrier A, Becce F, Farron A, Büchler P (2019) Automated CT bone segmentation using statistical shape modelling and local template matching. Comput Methods Biomech Biomed Engin 22:1303–1310. https://doi.org/10.1080/10255842.2019.1661391

32.

Damopoulos D, Glocker B, Zheng G (2018) Automatic localization of the lumbar vertebral landmarks in CT images with context features. In: Computational Methods and Clinical Applications in Musculoskeletal Imaging. Springer Cham, pp 59–71

33.

Forsberg D, Sjöblom E, Sunshine JL (2017) Detection and labeling of vertebrae in MR images using deep learning with clinical annotations as training data. J Digit Imaging 30:406–412. https://doi.org/10.1007/s10278-017-9945-xCrossRefPubMedPubMedCentral

34.

Payer C, Štern D, Bischof H, Urschler M (2016) Regressing heatmaps for multiple landmark localization using CNNs. In: Ourselin S, Joskowicz L, Sabuncu MR et al (eds) Medical image computing and computer-assisted intervention -- MICCAI 2016. Springer International Publishing, Cham, pp 230–238CrossRef

Title: Deep learning for the rapid automatic quantification and characterization of rotator cuff muscle degeneration from shoulder CT datasets
Authors: Elham Taghizadeh
Oskar Truffer
Fabio Becce
Sylvain Eminian
Stacey Gidoin
Alexandre Terrier
Alain Farron
Philippe Büchler
Publication date: 01-01-2021
Publisher: Springer Berlin Heidelberg
Keywords: Computed Tomography
Computed Tomography
Published in: European Radiology / Issue 1/2021
Print ISSN: 0938-7994
Electronic ISSN: 1432-1084
DOI: https://doi.org/10.1007/s00330-020-07070-7

Keynote webinar | Spotlight on medication adherence

Springer Medicine

Deep learning for the rapid automatic quantification and characterization of rotator cuff muscle degeneration from shoulder CT datasets

Abstract

Objectives

Methods

Results

Conclusions

Key Points

Keynote webinar | Spotlight on medication adherence

Springer Medicine

Abstract

Objectives

Methods

Results

Conclusions

Key Points

Please log in to get access to this content

Other articles of this Issue 1/2021

Comparison of contrast-enhanced videofluoroscopy to unenhanced dynamic MRI in minor patients following surgical correction of velopharyngeal dysfunction

Clinical evaluation of automated quantitative MRI reports for assessment of hippocampal sclerosis

Test-retest repeatability of a deep learning architecture in detecting and segmenting clinically significant prostate cancer on apparent diffusion coefficient (ADC) maps

MR image-based radiomics to differentiate type Ι and type ΙΙ epithelial ovarian cancers

A decade of radiomics research: are images really data or just patterns in the noise?

Utilization of radiomics to predict long-term outcome of magnetic resonance–guided focused ultrasound ablation therapy in adenomyosis