Functional Electrical Stimulation for Biceps Rehabilitation (2013-2016)
Eric J. Gonzalez, Ryan J. Downey, Courtney A. Rouse, and Warren E. Dixon. Undergraduate honors thesis research at the University of Florida Nonlinear Controls & Robotics Lab. Published in IEEE Transactions on Neural Systems and Rehabilitation Engineering.
Background
Functional electrical stimulation (FES) is the application of electrical stimulation to a muscle in order to elicit contraction and produce functional limb movement, such as grasping, walking, or cycling. FES is often used in physical rehabilitation to increase muscle strength, restore function, and improve range of motion in individuals that have suffered stroke or spinal cord injury. Traditionally, the implementation of FES involves the use of 2 surface electrodes per muscle group to deliver current to the muscle. It is known, however, that muscle length varies with limb flexion/extension and electrode position relative to the underlying muscle can impact contraction strength. Since functional limb flexion and extension through a wide range of motion is often desired, understanding the influence of limb flexion and stimulation site on muscle response to stimulation may lead to improved methods of delivering FES.
For my undergraduate honors thesis at the University of Florida in Prof. Warren Dixon's Nonlinear Controls and Robotics Lab, I designed and built a powered elbow-flexion testbed for upper arm FES and lead a study on the effects of elbow flexion and stimulation site on FES of the biceps. This work ultimately resulted in a first-authored journal publication in IEEE Trans. on Neural Systems and Rehabilitation Engineering.
Testbed Development
In order to test the effects of varying elbow flexion on FES, a device with the following functional capabilities was required:
The ability to collect kinematic information about the participant’s limb orientation (specifically their elbow joint angle and velocity)
The ability to measure the torque produced about the participant’s elbow axis when the biceps brachii is contracted, either volitionally or via external stimulation
The ability to resist or assist motion of the participant’s arm when desired
Considering these requirements, the testbed shown to the right was developed, consisting of a 12 VDC gear motor, optical encoder, and reaction torque sensor all in-line with the elbow axis of rotation. The crux of this design is the use of a hollow-shaft motor, which allowed the sensors to be placed compactly along the elbow axis via custom couplings with the motor. The system frame is made of 80-20 aluminum which allowed for modularity to accommodate different in arm sizes. The plates were manufactured from from water-jet aluminum. I also designed and manufactured (lathe, mill, & CNC) a custom hex shaft adapter that coupled the torque sensor to the motor. A RehaStim stimulator was used to provide stimulation to the muscle via surface electrodes.
Experiments
The aim of this work was to examine the relationship between elbow flexion angle, stimulation site, and stimulation-induced torque production in the biceps. We hypothesized that (1) the optimal stimulation site on the biceps varies with elbow flexion angle and (2) stimulation site on the biceps influences the flexion angle at which peak torque occurs. Electrodes were arranged in array over the bulk of the biceps. A protocol was written in MATLAB/Simulink to cycle the participant’s arm through a randomized sequence of flexion angles (0, 10, 20, …, 100 degrees). At each flexion angle, each of the six biceps electrodes received 1 second of stimulation (30 Hz, 25 mA, 100 μs) in a random order as elbow torque was recorded. The motor was locked such that stimulation-induced contractions of the biceps did not alter the flexion angle of the elbow (i.e., contractions were isometric).
We ran our study on 12 participants, and found that (1) flexion angle significantly influences the location of the optimal electrode within an electrode array and (2) stimulation site significantly influences the flexion angle at which contraction torque is maximal (i.e., peak-torque flexion angle). The results of this work motivate the use of kinematic-based electrode switching in FES tasks to maximize torque output.
