Development of MATLAB based EEG data acquisition for medical application

Zakaria, Zamry (2023) Development of MATLAB based EEG data acquisition for medical application. Project Report. Melaka, Malaysia, Universiti Teknikal Malaysia Melaka. (Submitted)

	Text (24 Pages) Development of MATLAB based EEG data acquisition for medical application.pdf - Submitted Version Download (930kB)
	Text (Full Text) Development of MATLAB based EEG data acquisition for medical application.pdf - Submitted Version Restricted to Registered users only Download (4MB)

Abstract

The electroencephalogram (EEG) is a recording made from the scalp of the brain's electrical activity. The objective of this study to develop of EEG data acquisition for medical application using MATLAB in this project using a medical application to detect epilepsy that is usually not detected in the early stages and is only detected when the patient has had a seizure because the seizure detection test requires a high cost., determine which kind of input data should be used to explore epilepsy seizures detection and achieve the best results for use with a neural network also to test and compare real brainwave data , obtained from EEG sensor by using Artificial Neural Network (ANN) using Bonn University dataset. By using hardware such as MindLink EEG Sensor, Bluetooth Module and Arduino Uno as a medium to take data from the patient's brain and send it to MATLAB software to analyze and get answers. In MATLAB software, the neural network Train and Test programs were developed to carry out operations using patternnet neuron value 25 after the experiment was conducted and found that this neuron value 25 can produce an accuracy of 83% which will produce an Artificial Neural Network (ANN) as a function test for assessing how well a trained ANN performs on a collection of test data. GUI (Graphical User Interface) in MATLAB are helpful for developing interactive tools and data visualization has been developed to make it easier for users to use the TEST program. The result reveals that all signals received from the EEG Sensor were successfully analyzed with a time rate according to the data value entered into the GUI Software. The results found that normal and abnormal results were successfully displayed on the GUI Software which took 20.86 seconds for 20 data, 50.73 seconds for 50 data, 1.42 minutes for 100 data, 17.10 minutes for 1000 data, 51.18 minutes for 3000 data and 4097 data took 67.18 minutes.

Actions (login required)

View Item