Introduction: Many patients suffering with autoimmune diseases that affect the joints, including rheumatoid arthritis, struggle with decreased gripping and coordination ability. This leads to increased pain, frustration, and emotional distress amidst what is already a challenging disease (Hewlett et al., 2012). By combining a wearable, nanocomposite strain gauge (Hyatt et al., 2010) and a cable-based actuator system mounted to a backpack frame (see Figure 1), an adaptive wrist posture device can respond to sudden changes in patient wrist position to help prevent the user dropping items and mitigate this struggle. Additionally, the device can detect directional changes towards flexion and extension and either assist or arrest motion to enable at-home physical therapy assistance. Recent related work has been done to create other assistive systems for the hand and wrist (Zhu et al., 2022), but none of these devices have sought to prevent sudden changes in wrist position and thereby prevent rapid motions that may induce pain or precipitate the unintended dropping of an object.
Materials and
Methods: The assistive wrist posture device is created from soft nanocomposite strain gauges attached to a commercial compression glove (see Figure 1). These strain gauges consist of nickel-coated carbon fiber and nickel nanostrand particles embedded in a silicone matrix and produce a variable resistance when their strain level changes. Mounted on the glove, they measure wrist strain by monitoring resistance output through an Arduino microcontroller. When the strain gauges experience a sudden change in strain (as is the case when the wrist jerks to initiate dropping an object unexpectedly), their viscoelastic properties cause them to produce a transient and dramatic spike in resistance (see Figure 2) that can be used to ascertain that sudden motion is occurring. The controller responds by actuation of a motor-driven cable system, which pulls on a cable line attached to the center of the patient’s hand to pull it back to a neutral position, providing support and preventing the unwanted rapid hand motion. Validation and calibration of the sensing portion of the device consisted of testing both directional and acceleration-based outputs. For the direction-based output, the wearer of the device moved smoothly between a neutral position, full flexion, a neutral position, and full extension repeatedly every 5 seconds (see Figure 3) while resistance data was recorded using MATLAB. To test the acceleration-based output, the wearer held a static position for 5 seconds before moving as rapidly as possible to full flexion or full extension (see Figure 2) while resistance data was again recorded in MATLAB.
Results, Conclusions, and Discussions: The recorded resistance data obtained during both the direction-based and acceleration-based trials shows the ability to differentiate between wrist positions and detect sudden movements. Thus, the wrist device can sense and respond to positional or acceleration input. From the acceleration-based data (see Figure 2), there are dramatic, transient spikes in resistance from the sensor whenever the wrist experiences a rapid change in position from neutral to flexion or extension. These rapid spikes have a very steep slope, which can be easily distinguished from the slope of normal, smooth data. This means that the device can be used to determine when rapid, unexpected motion occurs, which allows the actuation system to respond. Although the current version of the cable-pulling portion of the device lacks adequate motor speed and torque to prevent rapid motion, it does successfully demonstrate the concept of the device by correctly responding to the input. In the direction-based data (see Figure 3), there is an evident threshold value for resistance for the sensor on top (posterior wrist) and the sensor on the bottom (anterior wrist) of the glove. When the resistance value obtained from the bottom sensor exceeds its threshold value, the hand is in flexion. Likewise, when the resistance value of the top sensor exceeds its threshold, the hand is in extension. There is also a clear and repeated pattern in behavior for both top and bottom sensors displayed by the data when the hand motions of flexion and extension are repeated. This means that when the wrist posture device experiences smooth motion, it can be easily determined if the wrist is in flexion, extension, or a neutral position so that the cable-actuation system can appropriately respond. In combination, the two data results imply that the sensor-actuator device can be used to either arrest rapid motion (as is the case when an object is being dropped) or to track range of motion in physical therapy applications and assist the user to reach designated positions.
Acknowledgements (Optional):
References: 1. Hewlett, S., Sanderson, T., May, J., Alten, R., Bingham III, C.O., Cross, M., March, L., Pohl, C., Woodworth, T. and Bartlett, S.J., 2012. ‘I’m hurting, I want to kill myself’: rheumatoid arthritis flare is more than a high joint count—an international patient perspective on flare where medical help is sought. Rheumatology, 51(1), pp.69-76. 2. Hyatt, T., Fullwood, D., Bradshaw, R., Bowden, A. and Johnson, O., 2011. Nano-composite sensors for wide range measurement of ligament strain. In Experimental and Applied Mechanics, Volume 6: Proceedings of the 2010 Annual Conference on Experimental and Applied Mechanics (pp. 359-364). Springer New York. 3. Zhu, M., Biswas, S., Dinulescu, S.I., Kastor, N., Hawkes, E.W. and Visell, Y., 2022. Soft, wearable robotics and haptics: Technologies, trends, and emerging applications. Proceedings of the IEEE, 110(2), pp.246-272.
