A biplane 2d-3d registration protocol for quantifying glenohumeral kinematics following total shoulder arthroplasty

A better understanding of contact kinematics of the shoulder joint after total shoulder arthroplasty (TSA) may help identify causes of poor clinical outcomes following surgery [1]. In-vivo glenohumeral (GH) joint kinematics can be quantified from biplane fluoroscopic images using a 2D-3D image registration approach [2]. However, with some shoulder implants, tracking the glenoid component using fluoroscopy is challenging due to radiotransparency of the polyethylene component. An imaging protocol was proposed to overcome this limitation. The aim of this study was to describe the shoulder implant tracking accuracy using this protocol during dynamic biplane fluoroscopy. The overall RMS error of GH implant tracking was 1.72 mm in translation, and 0.96° in rotation, compared to radiostereometric analysis (RSA).