Author ORCID Identifier

https://orcid.org/0000-0003-1330-9560

Semester

Summer

Date of Graduation

2023

Document Type

Dissertation

Degree Type

PhD

College

School of Medicine

Department

Exercise Physiology

Committee Chair

Valeriya Gritsenko

Committee Member

Douglas Weber

Committee Member

Jean McCrory

Committee Member

Sergiy Yakovenko

Committee Member

Amelia Adcock

Abstract

Excellent motor control skills are necessary to live a high-quality life. Activities such as walking, getting dressed, and feeding yourself may seem mundane, but injuries to the neuromuscular system can render these tasks difficult or even impossible to accomplish without assistance. Statistics indicate that well over 100 million people are affected by diseases or injuries, such as stroke, Parkinson’s Disease, Multiple Sclerosis, Cerebral Palsy, peripheral nerve injury, spinal cord injury, and amputation, that negatively impact their motor abilities. This wide array of injuries presents a challenge to the medical field as optimal treatment paradigms are often difficult to implement due to a lack of availability of appropriate assessment tools, the inability for people to access the appropriate medical centers for treatment, or altogether gaps in technology for treating the underlying impairments causing the disability. Addressing each of these challenges will improve the treatment of movement impairments, provide more customized and continuous treatment to a larger number of patients, and advance rehabilitative and assistive device technology.

In my research, the key approach was to develop tools to assess and treat upper extremity movement impairment. In Chapter 2.1, I challenged a common biomechanical[GV1] modeling technique of the forearm. Comparing joint torque values through inverse dynamics simulation between two modeling platforms, I discovered that representing the forearm as a single cylindrical body was unable to capture the inertial parameters of a physiological forearm which is made up of two segments, the radius and ulna. I split the forearm segment into a proximal and distal segment, with the rationale being that the inertial parameters of the proximal segment could be tuned to those of the ulna and the inertial parameters of the distal segment could be tuned to those of the radius. Results showed a marked increase in joint torque calculation accuracy for those degrees of freedom that are affected by the inertial parameters of the radius and ulna. In Chapter 2.2, an inverse kinematic upper extremity model was developed for joint angle calculations from experimental motion capture data, with the rationale being that this would create an easy-to-use tool for clinicians and researchers to process their data. The results show accurate angle calculations when compared to algebraic solutions. Together, these chapters provide easy-to-use models and tools for processing movement assessment data. In Chapter 3.1, I developed a protocol to collect high-quality movement data in a virtual reality task that is used to assess hand function as part of a Box and Block Test. The goal of this chapter is to suggest a method to not only collect quality data in a research setting but can also be adapted for telehealth and at home movement assessment and rehabilitation. Results indicate that the data collected in this protocol are good and the virtual nature of this approach can make it a useful tool for continuous, data driven care in clinic or at home. In Chapter 3.2 I developed a high-density electromyography device for collecting motor unit action potentials of the arm. Traditional surface electromyography is limited by its ability to obtain signals from deep muscles and can also be time consuming to selectively place over appropriate muscles. With this high-density approach, muscle coverage is increased, placement time is decreased, and deep muscle activity can potentially be collected due to the high-density nature of the device[GV2] . Furthermore, the high-density electromyography device is built as a precursor to a high-density electromyography-electrical stimulation device for functional electrical stimulation. The customizable nature of the prototype in Chapter 3.2 allows for the implementation both recording and stimulating electrodes. Furthermore, signal results show that the electromyography data obtained from the device are of high quality and are correlated with gold standard surface electromyography sensors. One key factor in a device that can record and then stimulate based on the information from the recorded signals is an accurate movement intent decoder. High-quality movement decoders have been designed by closed-loop device controllers in the past, but they still struggle when the user interacts with objects of varying weight due to underlying alterations in muscle signals. In Chapter 4, I investigate this phenomenon by administering an experiment where participants perform a Box and Block Task with objects of 3 different weights, 0 kg, 0.02 kg, and 0.1 kg. Electromyography signals of the participants right arm were collected and co-contraction levels between antagonistic muscles were analyzed to uncover alterations in muscle forces and joint dynamics. Results indicated contraction differences between the conditions and also between movement stages (contraction levels before grabbing the block vs after touching the block) for each condition. This work builds a foundation for incorporating object weight estimates into closed-loop electromyography device movement decoders. Overall, we believe the chapters in this thesis provide a basis for increasing availability to movement assessment tools, increasing access to effective movement assessment and rehabilitation, and advance the medical device and technology field.

Download

Share

COinS