Pradana, Dio Alif and Nandang, I. and Munawar, S. Analisis Desain Sistem Klasifikasi Motor Imagery Menggunakan Metode WPT – CSP_CNN.
![[img]](http://repository.unik-kediri.ac.id/style/images/fileicons/text.png) |
Text
ANALISIS DESAIN SISTEM KLASIFIKASI MOTOR IMAGERY MENGGUNAKAN METODE WPT – CSP_CNN.doc Download (771kB) |
Abstract
This study is focused on proposed a new methods in EEG-based Brain Computer Interface (BCI) that can directly utilize brain signals to control external devices. Motor Imagery (MI) signal, which contains the imagination of a certain limb movement, is generally used in BCI. It does not need direct movement. The application of MI-EEG signal into BCI still has major problems because the patterns obtained for each recording can be different from one another even though they have the same type of motion. In this study, we utilize the Wavelet Packet Transform (WPT) method which is used to decompose the EEG signal into specifics sub-bands frequency and Common Spatial Pattern (CSP) as a spatial filter to increase the spatial resolution of the EEG signal. The Convolutional Neural Network (CNN) is then selected for training from the classifier. The results of this training will later be used to classify the movements of the given MI-EEG. We evaluate the model using dataset 2a from Brain-Computer Interface Competition (BCIC) IV. The results show the increases in the average accuracy increases 32%, Kappa up to 0.42, and decrease in Root Mean Square Error (RMSE) decreases up to 1.21 compared to only using CNN as the classifier. The performance of this algorithm also has a fairly good accuracy, RMSE, and Kappa value compared to other methods used previously in dataset 2a from BCIC IV. Keywords: MI-EEG, WPT, CSP, CNN, Spatial Resolution, Accuracy, Kappa, RMSE
| Item Type: | Article |
|---|---|
| Subjects: | T Technology > T Technology (General) |
| Divisions: | Fakultas Ilmu Kesehatan > Prodi Teknik Elektromedis D3 |
| Depositing User: | Unnamed user with email rio.mahardiko@gmail.com |
| Date Deposited: | 12 Aug 2023 03:46 |
| Last Modified: | 12 Aug 2023 03:46 |
| URI: | http://repository.unik-kediri.ac.id/id/eprint/651 |
Actions (login required)
 |
View Item |
