Get a New Perspective on EEG: Convolutional Neural Network Encoders for Parametric t-SNE
t-distributed stochastic neighbor embedding (t-SNE) is a method for reducing high-dimensional data to a low-dimensional representation, and is mostly used for visualizing data. In parametric t-SNE, a neural network learns to reproduce this mapping. When used for EEG analysis, the data are usually fi...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-03-01
|
| Series: | Brain Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3425/13/3/453 |
| _version_ | 1827750903437852672 |
|---|---|
| author | Mats Svantesson Håkan Olausson Anders Eklund Magnus Thordstein |
| author_facet | Mats Svantesson Håkan Olausson Anders Eklund Magnus Thordstein |
| author_sort | Mats Svantesson |
| collection | DOAJ |
| description | t-distributed stochastic neighbor embedding (t-SNE) is a method for reducing high-dimensional data to a low-dimensional representation, and is mostly used for visualizing data. In parametric t-SNE, a neural network learns to reproduce this mapping. When used for EEG analysis, the data are usually first transformed into a set of features, but it is not known which features are optimal. The principle of t-SNE was used to train convolutional neural network (CNN) encoders to learn to produce both a high- and a low-dimensional representation, eliminating the need for feature engineering. To evaluate the method, the Temple University EEG Corpus was used to create three datasets with distinct EEG characters: (1) wakefulness and sleep; (2) interictal epileptiform discharges; and (3) seizure activity. The CNN encoders produced low-dimensional representations of the datasets with a structure that conformed well to the EEG characters and generalized to new data. Compared to parametric t-SNE for either a short-time Fourier transform or wavelet representation of the datasets, the developed CNN encoders performed equally well in separating categories, as assessed by support vector machines. The CNN encoders generally produced a higher degree of clustering, both visually and in the number of clusters detected by k-means clustering. The developed principle is promising and could be further developed to create general tools for exploring relations in EEG data. |
| first_indexed | 2024-03-11T06:50:59Z |
| format | Article |
| id | doaj.art-d0790ba618484165a77aa05dc66676a0 |
| institution | Directory Open Access Journal |
| issn | 2076-3425 |
| language | English |
| last_indexed | 2024-03-11T06:50:59Z |
| publishDate | 2023-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Brain Sciences |
| spelling | doaj.art-d0790ba618484165a77aa05dc66676a02023-11-17T10:00:02ZengMDPI AGBrain Sciences2076-34252023-03-0113345310.3390/brainsci13030453Get a New Perspective on EEG: Convolutional Neural Network Encoders for Parametric t-SNEMats Svantesson0Håkan Olausson1Anders Eklund2Magnus Thordstein3Department of Clinical Neurophysiology, University Hospital of Linköping, 58185 Linköping, SwedenDepartment of Clinical Neurophysiology, University Hospital of Linköping, 58185 Linköping, SwedenCenter for Medical Image Science and Visualization, Linköping University, 58183 Linköping, SwedenDepartment of Clinical Neurophysiology, University Hospital of Linköping, 58185 Linköping, Swedent-distributed stochastic neighbor embedding (t-SNE) is a method for reducing high-dimensional data to a low-dimensional representation, and is mostly used for visualizing data. In parametric t-SNE, a neural network learns to reproduce this mapping. When used for EEG analysis, the data are usually first transformed into a set of features, but it is not known which features are optimal. The principle of t-SNE was used to train convolutional neural network (CNN) encoders to learn to produce both a high- and a low-dimensional representation, eliminating the need for feature engineering. To evaluate the method, the Temple University EEG Corpus was used to create three datasets with distinct EEG characters: (1) wakefulness and sleep; (2) interictal epileptiform discharges; and (3) seizure activity. The CNN encoders produced low-dimensional representations of the datasets with a structure that conformed well to the EEG characters and generalized to new data. Compared to parametric t-SNE for either a short-time Fourier transform or wavelet representation of the datasets, the developed CNN encoders performed equally well in separating categories, as assessed by support vector machines. The CNN encoders generally produced a higher degree of clustering, both visually and in the number of clusters detected by k-means clustering. The developed principle is promising and could be further developed to create general tools for exploring relations in EEG data.https://www.mdpi.com/2076-3425/13/3/453EEGdeep learningconvolutional neural networkst-SNEcategories |
| spellingShingle | Mats Svantesson Håkan Olausson Anders Eklund Magnus Thordstein Get a New Perspective on EEG: Convolutional Neural Network Encoders for Parametric t-SNE Brain Sciences EEG deep learning convolutional neural networks t-SNE categories |
| title | Get a New Perspective on EEG: Convolutional Neural Network Encoders for Parametric t-SNE |
| title_full | Get a New Perspective on EEG: Convolutional Neural Network Encoders for Parametric t-SNE |
| title_fullStr | Get a New Perspective on EEG: Convolutional Neural Network Encoders for Parametric t-SNE |
| title_full_unstemmed | Get a New Perspective on EEG: Convolutional Neural Network Encoders for Parametric t-SNE |
| title_short | Get a New Perspective on EEG: Convolutional Neural Network Encoders for Parametric t-SNE |
| title_sort | get a new perspective on eeg convolutional neural network encoders for parametric t sne |
| topic | EEG deep learning convolutional neural networks t-SNE categories |
| url | https://www.mdpi.com/2076-3425/13/3/453 |
| work_keys_str_mv | AT matssvantesson getanewperspectiveoneegconvolutionalneuralnetworkencodersforparametrictsne AT hakanolausson getanewperspectiveoneegconvolutionalneuralnetworkencodersforparametrictsne AT anderseklund getanewperspectiveoneegconvolutionalneuralnetworkencodersforparametrictsne AT magnusthordstein getanewperspectiveoneegconvolutionalneuralnetworkencodersforparametrictsne |