A Test Rig for the Analysis of the Knee Under Dynamic Motion Tasks1

Abstract

The study of the natural dynamic behavior of the knee is fundamental for the design of medical devices. In vivo tests allow the evaluation of the joint behavior on a patient, but require the use of complex (fluoroscopy), invasive (intracortical pins), or inaccurate (skin markers) techniques. In vitro tests overcome these problems, since allow all motion components to be precisely measured by trackers fixed to the bones. Moreover, medical devices can be tested during the design process and before the real implant on a patient. Conversely, loading conditions have to be reproduced on specimens and muscle activation patterns cannot be measured.Several devices for in vitro tests have been proposed to replicate the in vivo loading conditions of the knee, such as the Oxford knee rig (OKR) type [1] and robot-based systems [2]. Both types allow six degrees of freedom (6DOFs) to the femur–tibia relative motion, while external loads are applied to the joint. OKRs are simple, but cannot provide general loading conditions, unless their complexity increases. Robot-based systems require a complex closed-loop force control system, which increases costs and complexity. Furthermore, most of them were designed for industrial applications and are not optimized for the knee. For instance, several robots do not reach the high flexion angles required to test the knee.To overcome these limitations, a new test rig for measuring the knee joint motion under generic dynamic motion tasks was developed. The rig also allows estimation of the muscle forces required by a given task. The rig proved high accuracy and repeatability of measurements, despite its simplicity and low cost. In this paper, the main characteristics of the test rig are presented together with some recent experimental results.The rig was designed to evaluate both the loaded and unloaded behavior of the knee in a wide range of flexion. The whole frame of the rig is made up of two parts (Fig. 1): a base (5) and a portal (1). The base is fixed with respect to the lab, while the portal is connected to the base by a hinge (O). The femur (3) is connected to the portal through a fixation system (4) that allows specimen to be adjusted with 6DOFs with respect to the rig: Once the correct positioning is found, these 6DOFs are locked. The fixation system (4) allows also unmounting and precise repositioning of the femur. A ring (6) is fixed to the tibia (10). A generic load (force and moment) can be applied to the tibia by a system of wires connected to the ring. This system is composed of 12 wires that 2 × 2 act in the same direction. Each pair of wires belongs to a closed loop which includes a pneumatic actuator (8) fixed to the base. In each loop, only one wire is in tension at a time according to the applied load. The wires are arranged in a way that the load provided to the ring is practically fully decoupled: Actuators work in pairs to generate a force and a moment component along and about three orthogonal axes. Maximum force and moment are at least 2400 N and 100 N m over all directions, so that the most common daily activities can be replicated. The flexion angle is set by rotating the portal while keeping the axis of the tibia vertical: 135 deg of flexion can be reached. The loads are changed in real time with flexion to simulate a given task. The loading system is force-controlled: Except for the flexion, all rotations and displacements of the tibia are not constrained by the wires. Thus, the tibia is free to move and to reach the equilibrium configuration imposed by the wanted loads and by the knee anatomical structures.The rig is also equipped with a system that simulates the leg main muscles. By this system, muscle forces required by a given task can be experimentally estimated rather than imposed to the joint based on published data. In particular, a pneumatic actuator (2) (Fig. 1), fixed to the portal and connected to the quadriceps tendon, simulates the quadriceps. Conversely, flexor