Date of Submission

Fall 12-15-2019

Degree Type

Thesis

Degree Name

Master of Science in Computer Science (MSCS)

Department

Computer Science

Track

Big Data

Faculty Advisor

Chih-Cheng Hung

Chair

Coskun Cetinkaya

Committee Member

Chih-Cheng Hung

Committee Member

Junggab Son

Committee Member

Xiaohua Xu

Abstract

This paper attempts to answer the question of if it’s possible to produce a simple, quick, and accurate neural network for the use in upper-limb prosthetics. Through the implementation of convolutional and artificial neural networks and feature extraction on electromyographic data different possible architectures are examined with regards to processing time, complexity, and accuracy. It is found that the most accurate architecture is a multi-entry categorical cross entropy convolutional neural network with 100% accuracy. The issue is that it is also the slowest method requiring 9 minutes to run. The next best method found was a single-entry binary cross entropy convolutional neural network, which was able to reach an accuracy of about 95% in as little as 5 minutes. These time values, while being high for this research, are still a good deal faster than those found in some previous studies. These methods show promise in the popularization of machine learning algorithms in commercial prosthetics, which is something that is still uncommon.

Share

COinS