Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Convolutional Neural Network and Adversarial Autoencoder in EEG images classification

Albert Nasybullin, Semen Kurkin

Abstract

In this paper, we consider applying computer vision algorithms for the classification problem one faces in neuroscience during EEG data analysis. Our approach is to apply a combination of computer vision and neural network methods to solve human brain activity classification problems during hand movement. We pre-processed raw EEG signals and generated 2D EEG topograms. Later, we developed supervised and semi-supervised neural networks to classify different motor cortex activities.

Convolutional Neural Network and Adversarial Autoencoder in EEG images classification

Abstract

In this paper, we consider applying computer vision algorithms for the classification problem one faces in neuroscience during EEG data analysis. Our approach is to apply a combination of computer vision and neural network methods to solve human brain activity classification problems during hand movement. We pre-processed raw EEG signals and generated 2D EEG topograms. Later, we developed supervised and semi-supervised neural networks to classify different motor cortex activities.

Paper Structure

This paper contains 11 sections, 5 equations, 6 figures.

Table of Contents

  1. Introduction
  2. Data collection and data structure
  3. Task
  4. EEG data acquisition and preprocessing
  5. Data processing
  6. Finalized dataset
  7. Convolutional Neural Network (CNN)
  8. Results
  9. Semi-supervised Adversarial Autoencoder (AAE)
  10. Results
  11. Conclusion

Figures (6)

  • Figure 1: Left hand-related EEG topography in Mu frequency band.
  • Figure 2: Right hand-related EEG topography in Mu frequency band.
  • Figure 3: The CNN's accuracy is satisfying. The network reaches accuracy over 90$\%$even on small-scale data.
  • Figure 4: The Confusion Matrix of Adversarial Autoencoder. 1 = class "left hand", 0 = class "right hand".
  • Figure 5: Original Image after the augmentation is on a left. Generated Image is on a right. Step 0.
  • ...and 1 more figures