Comparative evaluation of scapular and humeral coordinate systems based on biomedical images of the glenohumeral joint.

Bio-imaging techniques represent a powerful tool for shoulder joint biomechanical analysis. However, the restricted field of view may prevent the acquisition of complete scapula and humerus bone models and hence limiting the applicability of standardized anatomical coordinate system (ACS) definitions. The aim of this study was to propose ACS definitions for both scapula and humerus which can be implemented when limited portions of the relevant bones are available. Magnetic resonance (MR) images of twenty right humeri and scapulae were acquired. The proposed ACSs were assessed in terms of (1) sensitivity to bone morphological variation, (2) intra - and inter - operator repeatability and (3) consistency with the anatomical cardinal directions. A comparison with alternative ACS definitions was also performed. Overall, our ACS scapular proposal and that presented in Kedgley and Dunning (2010) were found to be the least sensitive to the morphometric variability (mean angular absolute deviation lower than 8.3deg) and they were characterized by a high intra - and inter - operator repeatability (mean angular absolute deviation lower than 1.5deg). The humeral ACS proposal showed a morphometric variability similar to Amadi et al. (2009b) (mean angular absolute deviation lower than 8.3deg) but a higher reproducibility. The scapular and humeral ACS mean angular deviation from the reference anatomical cardinal directions were smaller than 15deg and 8.6deg, respectively. The proposed scapular and humeral ACS definitions are therefore suitable to be applied when a limited portion of the glenohumeral joint is available as it may occur in standard shoulder clinical exams.