A haptic surface scanning and machining parallel manipulator for registration-free bone resurfacing during arthroplasty

As orthopaedic implants become more sophisticated and complex, shaping the bone surface to match the implant becomes increasingly important. Accurate bone machining contributes to success of the implant and can potentially increase the implant's endurance and lifetime. Bone-mounted robots can eliminate intra-operative tracking errors during bone machining. However, registration errors still exist when aligning the pre-operative plan with the patient's anatomy because the plan is based on pre-operative images. Thus, a robotic system that is capable of executing both surface acquisition and bone machining in a single procedure and within the same coordinate system is preferable to the current state of the art. In this study we used a mini bone-attached robotic system (MBARS) with a haptic interface, to examine the accuracy of force-controlled surface acquisition of a femur model. The average distance error between acquired points and the surface model was less than 1 mm, which was further reduced by 5% with a compensation method which included curvature estimation of the reconstructed surface. An initial evaluation of in-situ bone machining based on these force-controlled scans yielded promising results.