3-D Dynamic and Static Control of the Knee and 3-D Actions of Individual Muscles

Information on the dynamic control (joint stiffness, viscosity and limb inertia) of the human knee joint is scarce in the literature, especially for actively contracting knee musculature and multiple axes. A joint driving device has been developed to apply small-amplitude random perturbations to the human knee about three orthogonal axes (flexion-extension, abduction-adduction, and internal-external rotation) and multiple angles with the subject maintaining various levels of muscle contraction. For example, it was found that joint stiffness and viscosity increased with muscle contraction substantially, while limb inertia was constant. Stiffness produced by the quadriceps was highest at 30° flexion and decreased with increasing or decreasing flexion angle, while knee flexors produced highest stiffness at 90° flexion. When knee flexion was < 60°, stiffness produced by the quadriceps was higher than that of the hamstrings and gastrocnemius at the same level of background muscle torque, while knee flexor muscles produced higher stiffnesses than the quadriceps at 90° flexion. Similar but less obvious trends were observed for joint viscosity. Passive joint stiffness at full knee extension was significantly higher than in more flexed positions. Surprisingly, as the knee joint musculature changed from relaxed to contracting at 50% MVC, system damping ratio remained at about 0.2. This outcome potentially simplifies neuromuscular control of the knee joint. In contrast, the natural undamped frequency increased more than twofold, potentially making the knee joint respond more quickly to the central nervous system commands. The approach described here provides us with a potentially valuable tool to quantify in vivo dynamic properties of normal and pathological human knee joints.

Fig. 1. Experimental setup of knee flexion-extension. The seat is adjusted in four degrees of freedom (DOF) to align the knee abduction axis with the motor shaft. The safety screws are used as mechanical stops to restrict the motor range of motion during the perturbation. A digital signal processor (DSP) controls the motor position. The DSP controller checks the joint position and torque signals at 2kHz and will shutdown the system if they are out of pre-specified ranges. A six-axis force sensor is mounted between the motor shaft and the aluminum beam. The short-leg cast is fixed to the aluminum beam through the two half-rings, and it can be adjusted and locked in four DOF to achieve appropriate alignment. The trunk was strapped to the backrest at 85° hip flexion. Fig. 2. Experimental setup of knee abduction-adduction. The seat is adjusted in four degrees of freedom (DOF) to align the knee abduction axis with the motor shaft. The safety screws are used as mechanical stops to restrict the motor range of motion during the perturbation. A digital signal processor (DSP) controls the motor position. The DSP controller checks the joint position and torque signals at 2kHz and will shutdown the system if they are out of pre-specified ranges. A six-axis force sensor is mounted between the motor shaft and the aluminum beam. The short-leg cast is fixed to the aluminum beam through the two half-rings, and it can be adjusted and locked in four DOF to achieve appropriate alignment. The trunk was strapped to the backrest at 85° hip flexion.
Fig. 3. Experimental setup for knee axial rotation. A six degrees of freedom goniometer is strapped to the thigh and leg to measure tibia motion relative to the femur in 3-D space. The seat is adjusted in four degrees of freedom (DOF) to align the tibial long axis with the motor shaft. The safety screws are used as mechanical stops to restrict the motor range of motion during the perturbation. A digital signal processor (DSP) controls the motor position. The DSP controller checks the joint position and torque signals at 2kHz and will shutdown the system if they are out of pre-specified ranges. A six-axis force sensor is mounted between the motor shaft and the L-shaped aluminum attachment. A short-leg cast is fixed to the aluminum attachment through the two half-rings, and it can be adjusted and locked in four DOF to achieve appropriate alignment. The thigh is strapped to the seat, and the trunk is strapped to the backrest at 85° hip flexion.
Mechanical Actions of Individual Muscles and Changes Caused by Musculoskeletal Injuries
Various orthopaedic injuries alter the mechanical actions of muscles and joint stability substantially. For example, rotator cuff tear is a major source of shoulder disability. However, it is not clear how rotator cuff tears change the directions in which an individual muscle crossing the glenohumeral joint rotate the humerus on the scapula in 3-D space and the torque-angle relationship of the muscle and how surgical reconstructions alter them again. The study is aimed at answering these clinically relevant questions through in vivo experiments on the patients. The study will help us understand the roles of individual shoulder muscles in performing functional tasks, gain insights into the effects and mechanisms of rotor cuff tears, and foster an in vivo means to compare different procedures for treating rotator cuff tears.