Study of anatomical landmark sampling error effect on motion measurement reproducibility for Orthopaedic physical examination assisting system

We have been developing the Orthopaedic physical examination assisting system using magnetic position tracker to quantitate knee laxity diagnosis. In this study, to investigate the influence of anatomical landmark sampling error for bone coordinate registration, reproducibility verification simulation was done. Firstly, the motion data of imitation bones was acquired using the system and set as a reference data. To get the error motion data, virtual error vectors were added on the reference sampling points and new motion data was recomputed. Evaluation was executed by comparison with the reference data and recomputed data. The evaluation results lead improvement methods of measurement reproducibility as follows. 1. It is important to point the landmarks proximal to the knee carefully because the contribution of the motion error is higher than the distal landmarks. 2. To reduce the registration error, thigh and shank receivers should be set close to the knee as possible. 3. If the allowance area of the sampling error is in 2.0 mm diameter circle, Diagnosis error using IKDC Knee examination form is approximately 20 %.