Signal2Image Modules in Deep Neural Networks for EEG Classification

TL;DR

The paper tackles EEG classification by converting signals into image-like inputs via Signal2Image modules to leverage image-based CNNs. It systematically compares four S2Is (signal-as-image, spectrogram, one-layer CNN, two-layer CNN) across multiple base models, using the UCI EEG seizure dataset segmented to 178-sample windows. Key findings show that a trainable one-layer CNN S2I yields the best or near-best performance across most base models (11/15), while non-trainable S2Is underperform and deeper S2Is provide limited gains. The work demonstrates the practicality of combining trainable S2Is with 1D base models for EEG classification and points to future directions in transfer learning and broader application to other physiological signals.

Abstract

Deep learning has revolutionized computer vision utilizing the increased availability of big data and the power of parallel computational units such as graphical processing units. The vast majority of deep learning research is conducted using images as training data, however the biomedical domain is rich in physiological signals that are used for diagnosis and prediction problems. It is still an open research question how to best utilize signals to train deep neural networks. In this paper we define the term Signal2Image (S2Is) as trainable or non-trainable prefix modules that convert signals, such as Electroencephalography (EEG), to image-like representations making them suitable for training image-based deep neural networks defined as `base models'. We compare the accuracy and time performance of four S2Is (`signal as image', spectrogram, one and two layer Convolutional Neural Networks (CNNs)) combined with a set of `base models' (LeNet, AlexNet, VGGnet, ResNet, DenseNet) along with the depth-wise and 1D variations of the latter. We also provide empirical evidence that the one layer CNN S2I performs better in eleven out of fifteen tested models than non-trainable S2Is for classifying EEG signals and we present visual comparisons of the outputs of the S2Is.

Figures (2)

• Figure 1: High level overview of a feed-forward pass of the combined methods. $x_i$ is the input, $m$ is the Signal2Image module, $b_{d}$ is the 1D or 2D architecture 'base model' for $d=1,2$ respectively and $\hat{y_i}$ is the predicted output. The names of the classes are depicted at the right along with the predictions for this example signal. The image between $m$ and $b_{d}$ depicts the output of the one layer CNN Signal2Image module, while the 'signal as image' and spectrogram have intermediate images as those depicted at the second and third row of Fig. \ref{['fig:signal2imageoutputs']}. Arrows denote the flow of the feed-forward pass. For the 1D architectures $m$ is omitted and no intermediate image is generated.
• Figure 2: Visualizations of the original signals and the outputs of the S2Is for each class. The x, y-axis of the first row are in µ V and time samples respectively. The x, y-axis of the rest of the subfigures denote spatial information, since we do not inform the 'base model' the concept of time along the x-axis or the concept of frequency along the y-axis. Higher pixel intensity denotes higher amplitude.