Identifying distinct scapular upward rotation patterns in individuals with upper trapezius–associated neck pain using unsupervised clustering approach

Background

Previous findings on scapular motion in upper trapezius (UT)-associated neck pain are inconsistent, and binary pain classification may overlook important movement characteristics.

Objectives

To identify distinct scapular upward rotation patterns in individuals with UT-associated neck pain using unsupervised clustering and to compare the resulting subgroups with healthy controls.

Design

Exploratory, cross-sectional design.

Method

Eighty-nine participants with UT-associated neck pain and 25 asymptomatic controls performed unilateral and bilateral scapular upward rotation. A smartphone camera recorded marker trajectories on the acromion process. Marker displacements along the x- and y-axes and the y/x ratio served as input features. K-means clustering was applied, and group differences were assessed.

Results/Findings

Cluster 1 showed restricted displacement and elevated y/x ratios; Cluster 2 displayed excessive displacement with normal y/x ratios. Control values lay between clusters, and pain intensity did not differ by cluster.

Conclusions

Smartphone-based clustering reveals hypomobile and hypermobile scapular phenotypes, supporting personalized assessment. This classification could potentially help clinicians who seek to assess how both excessive and restricted scapular motions may contribute to UT‐associated neck pain. Future studies should validate these clusters in larger samples and incorporate three-dimensional kinematics.