Shoulder Arthroplasty: Experimental Validation of Computer Generated Patient Specific Guides




Ellis, Adriel

Journal Title

Journal ISSN

Volume Title


University of Victoria


In a previous study done by J. Mackey, Patient Specific Guides (PSGs) were analyzed for their stability using a MATLAB tool called OrthoGrasp, and stability metrics were generated using Grasp Theory. Initial comparison tests were done with 3D printed glenoid models derived from CT scans, and PSGs altered from an original guide designed by an orthopaedic surgeon. This study sought to further compare an additional six glenoid models and their PSGs, created by M. Parmer, to their stability metrics generated by OrthoGrasp. The results showed that the average correlation between the two data sets was 0.57, lower than the 0.66 in the original study. Removing the two glenoid and PSG sets with extremely low correlation brought the average correlation to 0.70. On average, OrthoGrasp predicted the most stable PSG of the set 50% of the time. OrthoGrasp also predicted a 31.1% difference between the A-type and B-type guide models, where experimental data showed a 47% average difference between the two types of PSGs. These results show that OrthoGrasp could be a useful tool to help surgeons confirm the stability of a PSG, but likely can not be used as the sole tool for generating PSGs.



shoulder, arthroplasty, glenoid, patient specific guide, computer generated, validation