serval:BIB_CF8206F40600
Deep learning for the rapid automatic quantification and characterization of rotator cuff muscle degeneration from shoulder CT datasets.
10.1007/s00330-020-07070-7
000551008500001
32696257
Taghizadeh
E.
author
Truffer
O.
author
Becce
F.
author
Eminian
S.
author
Gidoin
S.
author
Terrier
A.
author
Farron
A.
author
Büchler
P.
author
article
2021-01
European radiology
1432-1084
0938-7994
journal
31
1
181-190
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.
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.
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 (R <sup>2</sup> = 0.87), fatty infiltration (R <sup>2</sup> = 0.91), and overall muscle degeneration (R <sup>2</sup> = 0.91). However, CNN-derived segmentations showed a higher variability in quantifying secondary bone formation (R <sup>2</sup> = 0.61) than human raters (R <sup>2</sup> = 0.87).
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.
• 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.
Adipose Tissue/diagnostic imaging
Adipose Tissue/pathology
Deep Learning
Humans
Muscular Atrophy/diagnostic imaging
Muscular Atrophy/pathology
Retrospective Studies
Rotator Cuff/diagnostic imaging
Rotator Cuff/pathology
Rotator Cuff Injuries
Shoulder
Tomography, X-Ray Computed
Computed tomography
Deep learning
Muscle atrophy
Rotator cuff
Sarcopenia
eng
60_published
true
peer-reviewed
Publication types: Journal Article
Publication Status: ppublish
University of Lausanne
mailto:serval_help@unil.ch
http://www.unil.ch/serval
http://serval.unil.ch/disclaimer
https://serval.unil.ch/notice/serval:BIB_CF8206F40600