ITR: Data-driven Human Knee Modeling for Expert Surgical Planning Systems

Western Pennsylvania Hospital Foundation, Pittsburgh PA

Investigators

Abstract

This work is developing a patient specific simulation tool that combines medical images and functional motion measurements to create a reliable biomechanical model suitable for pre- and intraoperative simulation of a patient's knee biomechanics after insertion of a implant. There currently exists no patient specific biomechanics simulator, and patient outcomes and quality of life should be improved through the ability to produce an ideal placement of medical devices and grafts. The work involves development of a novel knee biomechanical structure, based on parallel kinematic structures; automatic generation of patient-specific geometric models of joint anatomy from magnetic resonance images and deduction of soft-tissue information from motion capture data; and a surgical planner to determine the performance of a desired implant location, find the optimal placement of the implant, and simulate postoperative joint performance. The parallel kinematic description more faithfully represents the motion of a knee, which undergoes complex twist motions that current techniques attempt to model with simple revolute joints in a serial chain. Potential surgical outcomes are then modeled by inserting implants into the parallel structure and simulating the outcome, and optimizing the placement of the implant to maximally improve the positive results of the procedure. In addition to the technical achievements of this research, the broader outcome will include improved surgical results and quality of life for patients undergoing knee implant and graft procedures.