Automated quantification of glenoid bone defects using 3-dimensional measurements

Abstract

Background: Assessment of glenoid bone defects is important to select the optimal glenoid component design during shoulder arthroplasty planning and implantation. This study presents a fully automated method to describe glenoid bone loss using 3-dimensional measurements without the need for a healthy contralateral reference scapula.

Methods: The native shape of the glenoid is reconstructed by fitting a statistical shape model (SSM) of the scapula. The total vault loss percentage, local vault loss percentages, defect depth, defect area percentage, and subluxation distance and region are computed based on a comparison of the reconstructed and eroded glenoids. The method is evaluated by comparing its results with a contralateral bone-based reconstruction approach in a data set of 34 scapula and humerus pairs with unilateral glenoid bone defects.

Results: The SSM-based defect measurements deviated from the contralateral bone-based measurements with mean absolute differences of 5.5% in the total vault loss percentage, 4.5% to 8.0% in the local vault loss percentages, 1.9 mm in the defect depth, 14.8% in the defect area percentage, and 1.6 mm in the subluxation distance. The SSM-based method was statistically equivalent to the contralateral bone-based method for all parameters except the defect area percentage.

Conclusion: The presented method is able to automatically analyze glenoid bone defects using 3-dimensional measurements without the need for a healthy contralateral bone.

Keywords: Glenoid bone defects; automatization; defect measurements; preoperative planning; shoulder arthroplasty; statistical shape modeling.