Electronic Thesis and Dissertation Repository

Thesis Format

Integrated Article

Degree

Doctor of Philosophy

Program

Electrical and Computer Engineering

Supervisor

Dr. Ana Luisa Trejos

Abstract

Parkinson's disease (PD), the second most prevalent neurological disorder, is characterized by motor and non-motor symptoms including tremor. Wearable tremor suppression devices (WTSDs), have shown promising results for tremor reduction, using functional electrical stimulation (FES) or active actuators. However, results and study procedures are not consistent in the literature. Although stimulations below and above motor threshold can suppress tremor, the mechanisms of the first is still not clear, and the second might lead to muscle fatigue and discomfort over time. While active actuators have shown better results for tremor suppression, they are often heavy and bulky for daily use. Lastly, while many studies have analyzed characteristics of essential tremor (ET) under different circumstances, the difference between the pathophysiology of ET and PD suggests a need for better understanding of parkinsonian tremor characteristics.

To this end, a hybrid approach using electrical stimulation and mechanical suppression has been proposed. This system reduces the required motor torque and stimulation intensity for tremor suppression by using the other mechanism simultaneously, decreasing motor size, muscle fatigue, and discomfort. The hybrid approach improved tremor suppression by 12\% and reduced voluntary motion tracking error by 57\% compared to the FES-only approach in simulation. A case study showed that this approach could reduce the weight of a device with electric motors to about one-third of its initial weight.

Secondly, a systemic approach was proposed and tested to evaluate the effectiveness of various FES settings in parkinsonian tremor. Initially, individuals' tremor, sensory, and motor thresholds were evaluated. These measurements were used to generate stimulation combinations for tremor suppression at the wrist. Results showed that tremor suppression using FES highly depends on the tremor intensity, with a tremor power suppression ratio (TPSR) of 80.1 $\pm$ 2.9\% and 59.55 $\pm$ 2.9 \% for low to medium and higher tremor power intensities, respectively. Stimulations around the motor threshold showed an overall TPSR increase of 10\% and 4\% compared to below and above motor threshold stimulations, respectively.

Finally, approximate entropy (ApEn), frequency, power spectrum density, and magnitude of parkinsonian tremor were evaluated under different circumstances. An increase in ApEn from 0.74 $\pm$ 0.13 at baseline compared to 0.81 $\pm$ 0.22 with FES suppression aligns with previous studies, using surgery or medication for tremor suppression.

Understanding the effectiveness of different FES combinations on parkinsonian tremor can be used in further development of the hybrid approach, while findings of the last study are beneficial for the design of an adaptive controller.

Summary for Lay Audience

Many individuals across the world suffer from tremor. Tremor is one of the symptoms of many diseases including Parkinson's disease. Tremor has negative effects on people's lives and cannot be easily cured by medicine. As a potential solution, a worn robotic glove that can reduce tremor has become popular. These gloves reduce tremor by either twitching muscles or applying forces to the joints. However, currently they are heavy, bulky, or might result in muscle pain.

To improve these gloves, a new method was tested that reduces tremor by both twitching the muscles and applying force to the joints in a way that none of them become overwhelming and uncomfortable. Tests showed that using force and muscle twitch methods at the same time is better, since the same amount of tremor suppression can be achieved with less muscle twitch, and therefore, it causes less pain. Also, the tremor suppression requires less force on the joints, thus, less equipment on the glove, resulting in a lighter robotic glove.

Second, a test with individuals with Parkinson's disease was performed to find out how much muscle twitch can reduce tremor. Results showed that too little and too much muscle twitch have downsides, and an average value is better.

Lastly, parkinsonian tremor was studied for a better understanding of how tremor changes. These changes were compared at different times and situations. The results indicated that tremor not only changes in time but also is different while the individual is at rest or moving.

Overall, more tests are needed to build a good robotic glove. However, the results of this work showed that using both muscle twitch and force on the joints, while the level of muscle twitch is not very high or low, can be a good starting point for building a good robotic glove, as far as changes in the vibratory motion of tremor are being considered. Thus, the amount of twitch and force should be tuned based on the changes in tremor.

Download

Share

COinS