ULTRASONIC REGISTRATION OF KNEE ANATOMY TO MRI IMAGES

DESCRIPTION (provided by applicant): The proposed research will investigate the feasibility of using intra-operative ultrasound (US) images to noninvasively register the bone surfaces of the knee to preoperative magnetic resonance (MR) images. A surgical navigation system, KneeNav, is being developed by CASurgica for intraoperatively planning the proper location of ligament attachment sites and drill tunnel locations and then guiding the surgeon to execute the plan. Despite agreement on the correct points of insertion, great variability exists in tunnel placement among surgeons and the rate of misplaced tunnels in ACL reconstruction surgery has been reported to be between 10-40 percent. Malpositioning of the bone tunnels is the main reason for revision surgery. The proposed research would increase the accuracy of the procedure versus current videoscopic techniques or versus competitive image guidance systems. The research plan is to establish a "gold standard" for registration accuracy using reconstructed CT scans, point-based surface matching algorithms, and optical tracking. Reconstructed MR models will be substituted for CT models and the accuracy reassessed. The US probe will then be calibrated and US surface collection will then be substituted for point based collection and the accuracy reassessed. Finally, registration of US directly to MR without reconstruction will be assessed. PROPOSED COMMERCIAL APPLICATION: Not Available.