Deep-Learning-Based Automatic Selection of Fewest Channels for Brain-Machine Interfaces
Abstract : Due to the development of convenient brain-machine interfaces (BMIs), the automatic selection of a minimum channel (electrode) set has attracted increasing interest because the decrease in the number of channels increases the efficiency of BMIs. This study proposes a deep-learning-based technique to automatically search for the minimum number of channels applicable to general BMI paradigms using a compact convolutional neural network for electroencephalography (EEG)-based BMIs. For verification, three types of BMI paradigms are assessed: 1) the typical P300 auditory oddball; 2) the new top-down steady-state visually evoked potential; and 3) the endogenous motor imagery. We observe that the optimized minimal EEG-channel sets are automatically selected in all three cases. Their decoding accuracies using the minimal channels are statistically equivalent to (or even higher than) those based on all channels. The brain areas of the selected channel set are neurophysiologically interpretable for all of these cognitive task paradigms. This study shows that the minimal EEG channel set can be automatically selected, irrespective of the types of BMI paradigms or EEG input features using a deep-learning approach, which also contributes to their portability.
? Brain-computer interface (BCI) is an emerging area of research with enormous scope in medical as well as non-medical applications.
? Most of the existing machine learning studies focus on static data and cannot classify the dynamic changes of brain signals accurately for practical uses.
? There exist many types of feature selection algorithms like filters, wrapper etc.
? The design of the experiment, data acquisition, and training requires a high degree of organization and timing.
? Developing a large database is quite tedious and expensive. The variability of the brain signals produced across subjects is also quite high.
? The security problem can be divided into identi?cation (also called recognition) and authentication (also called veri?cation) aspects.
? To solve the problem that the intracortical signals are expensive to collect, the authors also proposed a deep neural architecture to aiming at mapping the scalp signals to pseudo-intracranial brain signals.
? The former generally is a multi-class classification problem, and its target is to recognize the identity of the test-person.
? Some papers only attempt to classify the user’s emotional state into a binary (positive/negative) or three categories (positive, neutral, and negative) problem and recognize them by deep learning algorithms.
• The proposed Reinforced CNN selects the best attention area that leads to the highest classification accuracy using a non-linear reward function to encourage the model.
• The proposed models filter the output of CNN using Discrete Wavelet Transform (DWT) with Coiflet wavelet mother signal.
• Others argue for attributing the good performance to smoothing the objective function, while others propose that lengthdirection decoupling is the reason behind its effectiveness.
• Some studies propose new network structures that mix CNN with representation algorithms for feature extraction and classification.
• To reduce feature dimensions, a new approach is proposed to select channels and features that represent the highest emotional state.
? These represent but a small sample of the entire set of neurons in this limited region, as spiking can only be detected by microelectrodes closely approximated to a neuron developed a high performance BCI system for communication of ALS patients.
? The wider frequency bands take substantial information from functional areas of a brain (e.g., motor and language) and thus can be used to train a higher-performance BCI system.
? Moreover, the performance of ERD/ERS among users is quite variable, and the accuracy is not very high.
? They have taken advantage of MLP for detecting hierarchical features and LSTM for sequential data learning to optimize classification performance with single-channel recordings.
|