Objective analysis of partial three-dimensional rotator cuff muscle volume and fat infiltration across ages and sex from clinical MRI scans.

Objective analysis of rotator cuff (RC) atrophy and fatty infiltration (FI) from clinical MRI is limited by qualitative measures and variation in scapular coverage. The goals of this study were to: develop/evaluate a method to quantify RC muscle size, atrophy, and FI from clinical MRIs (with typical lateral only coverage) and then quantify the effects of age and sex on RC muscle. To develop the method, 47 full scapula coverage CTs with matching clinical MRIs were used to: correct for variation in scan capture, and ensure impactful information of the RC is measured. Utilizing this methodology and automated artificial intelligence, 170 healthy clinical shoulder MRIs of varying age and sex were segmented, and each RC muscle's size, relative contribution, and FI as a function of scapula location were quantified. A two-way ANOVA was used to examine the effect of age and sex on RC musculature. The analysis revealed significant (p < 0.05): decreases in size of the supraspinatus, teres minor, and subscapularis with age; decreased supraspinatus and increased infraspinatus relative contribution with age; and increased FI in the infraspinatus with age and in females. This study demonstrated that clinically obtained MRIs can be utilized for automatic 3D analysis of the RC. This method is not susceptible to coverage variation or patient size. Application of methodology in a healthy population revealed differences in RC musculature across ages and FI level between sexes. This large database can be used to reference expected muscle characteristics as a function of scapula location and could eventually be used in conjunction with the proposed methodology for analysis in patient populations.

A Deep Learning Algorithm for Automatic 3D Segmentation of Rotator Cuff Muscle and Fat from Clinical MRI Scans.

The authors aimed to develop and validate an automated artificial intelligence (AI) algorithm for three-dimensional (3D) segmentation of all four rotator cuff (RC) muscles to quantify intramuscular fat infiltration (FI) and individual muscle volume. The dataset included retrospectively collected RC MRI scans in 232 patients (63 with normal RCs, 169 with RC tears). A two-stage AI model was developed to segment all RC muscles and their FI in each stage. For comparison, single-stage and Otsu filtering models were created. Using the two-stage model, segmentation performance demonstrated high Dice scores (mean, 0.92 ± 0.14 [SD]), low volume errors (mean, 5.72% ± 9.23), and low FI errors (mean, 1.54% ± 2.79) when validated in 30 scans. There was a significant correlation between the 3D FI in the RC tear scans with a Goutallier grade (ρ = 0.53, P < .001) and FI found from a single two-dimensional (2D) section (all muscles, ρ > 0.70; P < .001). However, Bland-Altman analysis of the 3D compared with the 2D analyses of FI demonstrated a proportional bias (all muscles, P < .001). Compared with Goutallier classification or single-image quantification, the AI method allowed for more variability in images and led to objective separate quantifications of muscle volume and FI in all RC muscles. Keywords: Rotator Cuff, Artificial Intelligence, Segmentation, Fat Infiltration, Muscle Volume, MRI, Shoulder Supplemental material is available for this article. © RSNA, 2023.