Key result
A two-stage AI model for 3D segmentation of rotator cuff muscles achieved high Dice scores (mean 0.92) and correlated significantly with Goutallier grade (ρ = 0.53, P < .001).
Why the study?
The authors aimed to develop and validate an automated artificial intelligence algorithm for 3D segmentation of all four rotator cuff muscles to quantify intramuscular fat infiltration and individual muscle volume.
Effect estimate: ρ = 0.53
p-value: p=< .001
An automated two-stage AI algorithm enables objective 3D quantification of rotator cuff muscle volume and fat infiltration from clinical MRI scans, outperforming traditional 2D methods.
May aid rotator cuff fat quantification on MRI; hypothesis-generating pending prospective validation.
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.
No takes yet. Share an insight, caveat, or question.
Riem et al. (2023) studied Rotator cuff tears (n=232). Two-stage AI model for 3D segmentation vs. Single-stage model, Otsu filtering, Goutallier classification, and single 2D section was evaluated on Segmentation performance (Dice scores) and correlation of 3D fat infiltration with Goutallier grade (ρ = 0.53, p=< .001). A two-stage AI model for 3D segmentation of rotator cuff muscles achieved high Dice scores (mean 0.92) and correlated significantly with Goutallier grade (ρ = 0.53, P < .001).
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: