Characterization of robotic system passive path repeatability during specimen removal and reinstallation for in vitro knee joint testing.

Robotic testing systems are commonly utilized for the study of orthopaedic biomechanics. Quantification of system error is essential for reliable use of robotic systems. Therefore, the purpose of this study was to quantify a 6-DOF robotic system's repeatability during knee biomechanical testing and characterize the error induced in passive path repeatability by removing and reinstalling the knee. We hypothesized removing and reinstalling the knee would substantially alter passive path repeatability. Testing was performed on four fresh-frozen cadaver knees. To determine repeatability and reproducibility, the passive path was collected three times per knee following the initial setup (intra-setup), and a single time following two subsequent re-setups (inter-setup). Repeatability was calculated as root mean square error. The intra-setup passive path had a position repeatability of 0.23 mm. In contrast, inter-setup passive paths had a position repeatability of 0.89 mm. When a previously collected passive path was replayed following re-setup of the knee, resultant total force repeatability across the passive path increased to 28.2N (6.4N medial-lateral, 25.4N proximal-distal, and 10.5 N anterior-posterior). This study demonstrated that removal and re-setup of a knee can have substantial, clinically significant changes on our system's repeatability and ultimately, accuracy of the reported results.