Preoperative radiomic biomarkers reflect functional shoulder impairment in rotator cuff tears: a structure-function analysis.

Introduction: Rotator cuff tears (RCTs) are a major cause of shoulder dysfunction and functional limitation. Although magnetic resonance imaging (MRI) is fundamental for assessing tendon integrity, it provides a static representation that may not fully explain functional impairment. Optoelectronic motion capture systems enable quantitative assessment of three-dimensional (3D) shoulder kinematics. The aim of this study was to investigate the relationship between preoperative MRI-derived radiomic features and shoulder kinematic parameters in patients with supraspinatus tendon tears. Methods: A total of 73 patients with RCTs underwent preoperative MRI and 3D shoulder kinematic assessment during four dynamic movements. Radiomic features were extracted from MRI images, while kinematic parameters were computed from motion capture data. Correlation analyses were conducted on the entire patient cohort and after stratification by the American Shoulder and Elbow Surgeons (ASES) score to further investigate whether specific clinical characteristics influenced these relationships. Results: No significant associations were observed in the full cohort. In contrast, multiple moderate radiomic-kinematic correlations emerged in ASES-based subgroups. Both positive and negative associations were identified between shoulder kinematic parameters and radiomic feature categories, predominantly involving texture- and intensity-based features, with additional contributions from shape and general descriptors. Discussion: By integrating radiomics with dynamic motion analysis, this study shows that MRI-derived tissue-level alterations are associated with real-world functional shoulder performance, reinforcing the value of combining imaging-based biomarkers with functional assessments for a more comprehensive evaluation of shoulder pathology.