Emg features extraction and performance indicators for upper limb prosthetics

Eissa, Ali Huessin Ahmed (2018) Emg features extraction and performance indicators for upper limb prosthetics. Master's Internship Report, Biomedical Engineering.

Preview	Text MBME_2018_EissaAHA.pdf Download (1MB) | Preview
	Text toestemming.pdf Restricted to Registered users only Download (153kB)

Abstract

In a test carried out by members of project INPUT, EMG readings from surface electrodes were collected from 38 able bodied participants. The participants were given movement tasks to train the pattern recognition algorithm used in the project, and to test the ability of the algorithm to classify the movements from the extracted EMG signal, using the motion test. In this project, with the help of the platform BioPatRec, data collected from participants can be further studied. This could be done by extracting different features, and checking the classifiers interaction with these features, as well as the possibility to use different classifiers. However, the data was collected and stored in a format that was incompatible with the required data format for BioPatRec. Therefore, the primary goal of this study is to use Matlab to manipulate the structure of the data set used in project INPUT. This is expected to allow the use of the data on the platform BioPatRec. The adjusted data can then be used to check for a possible correlation between the feature data sets and the ability of able bodied users to use the classifier algorithm. With the help of some statistical analysis, a further understanding of the effect of feature changes on performance could be acquired.

Item Type:	Thesis (Master's Internship Report)
Supervisor name:	Verkerke, G.J.
Degree programme:	Biomedical Engineering
Thesis type:	Master's Internship Report
Language:	English
Date Deposited:	27 Nov 2018
Last Modified:	03 Dec 2018 13:10
URI:	https://fse.studenttheses.ub.rug.nl/id/eprint/18850

Actions (login required)

View Item