CWT Based Transfer Learning for Motor Imagery Classification for Brain computer Interfaces

EEG-BCI-based motor imagery classification using double attention convolutional network

This article aims to improve and diversify signal processing techniques to execute a brain-computer interface (BCI) based on neurological phenomena observed when performing motor tasks using motor imagery (MI). The noise present in the original data, such as intermodulation noise, crosstalk, and other unwanted noise, is removed by Modify Least Mean Square (M-LMS) in the pre-processing stage. Traditional LMSs were unable to extract all the noise from the images. After pre-processing, the required features, such as statistical features, entropy features, etc., were extracted using Common Spatial Pattern (CSP) and Pearson's Correlation Coefficient (PCC) instead of the traditional single feature extraction model. The arithmetic optimization algorithm cannot select the features accurately and fails to reduce the feature dimensionality of the data. Thus, an Extended Arithmetic operation optimization (ExAo) algorithm is used to select the most significant attributes from the extracted features. The proposed model uses Double Attention Convolutional Neural Networks (DAttnConvNet) to classify the types of EEG signals based on optimal feature selection. Here, the attention mechanism is used to select and optimize the features to improve the classification accuracy and efficiency of the model. In EEG motor imagery datasets, the proposed model has been analyzed under class, which obtained an accuracy of 99.98% in class Baseline (B), 99.82% in class Imagined movement of a right fist (R) and 99.61% in class Imagined movement of both fists (RL). In the EEG dataset, the proposed model can obtain a high accuracy of 97.94% compared to EEG datasets of other models.

A multi-feature fusion graph attention network for decoding motor imagery intention in spinal cord injury patients

Objective.Electroencephalogram (EEG) signals exhibit temporal-frequency-spatial multi-domain feature, and due to the nonplanar nature of the brain surface, the electrode distributions follow non-Euclidean topology. To fully resolve the EEG signals, this study proposes a temporal-frequency-spatial multi-domain feature fusion graph attention network (GAT) for motor imagery (MI) intention recognition in spinal cord injury (SCI) patients.Approach.The proposed model uses phase-locked value (PLV) to extract spatial phase connectivity information between EEG channels and continuous wavelet transform to extract valid EEG information in the time-frequency domain. It then models as a graph data structure containing multi-domain information. The gated recurrent unit and GAT learn EEG's dynamic temporal-spatial information. Finally, the fully connected layer outputs the MI intention recognition results.Main results.After 10 times 10-fold cross-validation, the proposed model can achieve an average accuracy of 95.82%. Furthermore, this study analyses the event-related desynchronization/event-related synchronization and PLV brain network to explore the brain activity of SCI patients during MI.Significance.This study confirms the potential of the proposed model in terms of EEG decoding performance and provides a reference for the mechanism of neural activity in SCI patients.

FDCN-C: A deep learning model based on frequency enhancement, deformable convolution network, and crop module for electroencephalography motor imagery classification

Motor imagery (MI)-electroencephalography (EEG) decoding plays an important role in brain-computer interface (BCI), which enables motor-disabled patients to communicate with external world via manipulating smart equipment. Currently, deep learning (DL)-based methods are popular for EEG decoding. Whereas the utilization efficiency of EEG features in frequency and temporal domain is not sufficient, which results in poor MI classification performance. To address this issue, an EEG-based MI classification model based on a frequency enhancement module, a deformable convolutional network, and a crop module (FDCN-C) is proposed. Firstly, the frequency enhancement module is innovatively designed to address the issue of extracting frequency information. It utilizes convolution kernels at continuous time scales to extract features across different frequency bands. These features are screened by calculating attention and integrated into the original EEG data. Secondly, for temporal feature extraction, a deformable convolution network is employed to enhance feature extraction capabilities, utilizing offset parameters to modulate the convolution kernel size. In spatial domain, a one-dimensional convolution layer is designed to integrate all channel information. Finally, a dilated convolution is used to form a crop classification module, wherein the diverse receptive fields of the EEG data are computed multiple times. Two public datasets are employed to verify the proposed FDCN-C model, the classification accuracy obtained from the proposed model is greater than that of state-of-the-art methods. The model's accuracy has improved by 14.01% compared to the baseline model, and the ablation study has confirmed the effectiveness of each module in the model.

A class alignment network based on self-attention for cross-subject EEG classification

Due to the inherent variability in EEG signals across different individuals, domain adaptation and adversarial learning strategies are being progressively utilized to develop subject-specific classification models by leveraging data from other subjects. These approaches primarily focus on domain alignment and tend to overlook the critical task-specific class boundaries. This oversight can result in weak correlation between the extracted features and categories. To address these challenges, we propose a novel model that uses the known information from multiple subjects to bolster EEG classification for an individual subject through adversarial learning strategies. Our method begins by extracting both shallow and attention-driven deep features from EEG signals. Subsequently, we employ a class discriminator to encourage the same-class features from different domains to converge while ensuring that the different-class features diverge. This is achieved using our proposed discrimination loss function, which is designed to minimize the feature distance for samples of the same class across different domains while maximizing it for those from different classes. Additionally, our model incorporates two parallel classifiers that are harmonious yet distinct and jointly contribute to decision-making. Extensive testing on two publicly available EEG datasets has validated our model's efficacy and superiority.

A Cross-Attention-Based Class Alignment Network for Cross-Subject EEG Classification in a Heterogeneous Space

BACKGROUND

Domain adaptation (DA) techniques have emerged as a pivotal strategy in addressing the challenges of cross-subject classification. However, traditional DA methods are inherently limited by the assumption of a homogeneous space, requiring that the source and target domains share identical feature dimensions and label sets, which is often impractical in real-world applications. Therefore, effectively addressing the challenge of EEG classification under heterogeneous spaces has emerged as a crucial research topic.

METHODS

We present a comprehensive framework that addresses the challenges of heterogeneous spaces by implementing a cross-domain class alignment strategy. We innovatively construct a cross-encoder to effectively capture the intricate dependencies between data across domains. We also introduce a tailored class discriminator accompanied by a corresponding loss function. By optimizing the loss function, we facilitate the aggregation of features with corresponding classes between the source and target domains, while ensuring that features from non-corresponding classes are dispersed.

RESULTS

Extensive experiments were conducted on two publicly available EEG datasets. Compared to advanced methods that combine label alignment with transfer learning, our method demonstrated superior performance across five heterogeneous space scenarios. Notably, in four heterogeneous label space scenarios, our method outperformed the advanced methods by an average of 7.8%. Moreover, in complex scenarios involving both heterogeneous label spaces and heterogeneous feature spaces, our method outperformed the state-of-the-art methods by an average of 4.1%.

CONCLUSIONS

This paper presents an efficient model for cross-subject EEG classification under heterogeneous spaces, which significantly addresses the challenges of EEG classification within heterogeneous spaces, thereby opening up new perspectives and avenues for research in related fields.

Emotion recognition with reduced channels using CWT based EEG feature representation and a CNN classifier

Objective.Although emotion recognition has been studied for decades, a more accurate classification method that requires less computing is still needed. At present, in many studies, EEG features are extracted from all channels to recognize emotional states, however, there is a lack of an efficient feature domain that improves classification performance and reduces the number of EEG channels.Approach.In this study, a continuous wavelet transform (CWT)-based feature representation of multi-channel EEG data is proposed for automatic emotion recognition. In the proposed feature, the time-frequency domain information is preserved by using CWT coefficients. For a particular EEG channel, each CWT coefficient is mapped into a strength-to-entropy component ratio to obtain a 2D representation. Finally, a 2D feature matrix, namely CEF2D, is created by concatenating these representations from different channels and fed into a deep convolutional neural network architecture. Based on the CWT domain energy-to-entropy ratio, effective channel and CWT scale selection schemes are also proposed to reduce computational complexity.Main results.Compared with previous studies, the results of this study show that valence and arousal classification accuracy has improved in both 3-class and 2-class cases. For the 2-class problem, the average accuracies obtained for valence and arousal dimensions are 98.83% and 98.95%, respectively, and for the 3-class, the accuracies are 98.25% and 98.68%, respectively.Significance.Our findings show that the entropy-based feature of EEG data in the CWT domain is effective for emotion recognition. Utilizing the proposed feature domain, an effective channel selection method can reduce computational complexity.

Classification of Motor Imagery EEG signals using high resolution time-frequency representations and convolutional neural network

A Motor Imagery (MI) based Brain Computer Interface (BCI) system aims to provide neuro-rehabilitation for the motor disabled people and patients with brain injuries (e.g., stroke patients) etc. The aim of this work is to classify the left and right hand MI tasks by utilizing the occurrence of event related desynchronization and synchronization (ERD\ERS) in the Electroencephalogram (EEG) during these tasks. This study proposes to use a set of Complex Morlet Wavelets (CMW) having frequency dependent widths to generate high-resolution time-frequency representations (TFR) of the MI EEG signals present in the channels C3 and C4. A novel method for the selection of the value of number of cycles relative to the center frequency of the CMW is studied here for extracting the MI task features. The generated TFRs are given as input to a Convolutional neural network (CNN) for classifying them into left or right hand MI tasks. The proposed framework attains a classification accuracy of 82.2% on the BCI Competition IV dataset 2a, showing that the TFRs generated in this work give a higher classification accuracy than the baseline methods and other existing algorithms.

Non-Destructive Characterization of Cured-in-Place Pipe Defects

Sewage and water networks are crucial infrastructures of modern urban society. The uninterrupted functionality of these networks is paramount, necessitating regular maintenance and rehabilitation. In densely populated urban areas, trenchless methods, particularly those employing cured-in-place pipe technology, have emerged as the most cost-efficient approach for network rehabilitation. Common diagnostic methods for assessing pipe conditions, whether original or retrofitted with-cured-in-place pipes, typically include camera examination or laser scans, and are limited in material characterization. This study introduces three innovative methods for characterizing critical aspects of pipe conditions. The impact-echo method, ground-penetrating radar, and impedance spectroscopy address the challenges posed by polymer liners and offer enhanced accuracy in defect detection. These methods enable the characterization of delamination, identification of caverns behind cured-in-place pipes, and evaluation of overall pipe health. A machine learning algorithm using deep learning on images acquired from impact-echo signals using continuous wavelet transformation is presented to characterize defects. The aim is to compare traditional machine learning and deep learning methods to characterize selected pipe defects. The measurement conducted with ground-penetrating radar is depicted, employing a heuristic algorithm to estimate caverns behind the tested polymer composites. This study also presents results obtained through impedance spectroscopy, employed to characterize the delamination of polymer liners caused by uneven curing. A comparative analysis of these methods is conducted, assessing the accuracy by comparing the known positions of defects with their predicted characteristics based on laboratory measurements.

Development of hybrid feature learner model integrating FDOSM for golden subject identification in motor imagery

Brain-computer interfaces (BCIs) based on motor imagery (MI) face challenges due to the complex nature of brain activity, nonstationary and high-dimensional properties, and individual variations in motor behaviour. The identification of a consistent "golden subject" in MI-based BCIs remains an open challenge, complicated by multiple evaluation metrics and conflicting trade-offs, presenting complex Multi-Criteria Decision Making (MCDM) problems. This study proposes a hybrid brain signal decoding model called Hybrid Adaboost Feature Learner (HAFL), which combines feature extraction and classification using VGG-19, STFT, and Adaboost classifier. The model is validated using a pre-recorded MI-EEG dataset from the BCI competition at Graz University. The fuzzy decision-making framework is integrated with HAFL to allocate a golden subject for MI-BCI applications through the Golden Subject Decision Matrix (GSDM) and the Fuzzy Decision by Opinion Score Method (FDOSM). The effectiveness of the HAFL model in addressing inter-subject variability in EEG-based MI-BCI is evaluated using an MI-EEG dataset involving nine subjects. Comparing subject performance fairly is challenging due to complexity variations, but the FDOSM method provides valuable insights. Through FDOSM-based External Group Aggregation (EGA), subject S5 achieves the highest score of 2.900, identified as the most promising golden subject for subject-to-subject transfer learning. The proposed methodology is compared against other benchmark studies from various key perspectives and exhibits significant novelty in several aspects. The findings contribute to the development of more robust and effective BCI systems, paving the way for advancements in subject-to-subject transfer learning for BCI-MI applications.

MBGA-Net: A multi-branch graph adaptive network for individualized motor imagery EEG classification

BACKGROUND AND OBJECTIVE

The development of deep learning has led to significant improvements in the decoding accuracy of Motor Imagery (MI) EEG signal classification. However, current models are inadequate in ensuring high levels of classification accuracy for an individual. Since MI EEG data is primarily used in medical rehabilitation and intelligent control, it is crucial to ensure that each individual's EEG signal is recognized with precision.

METHODS

We propose a multi-branch graph adaptive network (MBGA-Net), which matches each individual EEG signal with a suitable time-frequency domain processing method based on spatio-temporal domain features. We then feed the signal into the relevant model branch using an adaptive technique. Through an enhanced attention mechanism and deep convolutional method with residual connectivity, each model branch more effectively harvests the features of the related format data.

RESULTS

We validate the proposed model using the BCI Competition IV dataset 2a and dataset 2b. On dataset 2a, the average accuracy and kappa values are 87.49% and 0.83, respectively. The standard deviation of individual kappa values is only 0.08. For dataset 2b, the average classification accuracies obtained by feeding the data into the three branches of MBGA-Net are 85.71%, 85.83%, and 86.99%, respectively.

CONCLUSIONS

The experimental results demonstrate that MBGA-Net could effectively perform the classification task of motor imagery EEG signals, and it exhibits strong generalization performance. The proposed adaptive matching technique enhances the classification accuracy of each individual, which is beneficial for the practical application of EEG classification.

Portable deep-learning decoder for motor imaginary EEG signals based on a novel compact convolutional neural network incorporating spatial-attention mechanism

Due to high computational requirements, deep-learning decoders for motor imaginary (MI) electroencephalography (EEG) signals are usually implemented on bulky and heavy computing devices that are inconvenient for physical actions. To date, the application of deep-learning techniques in independent portable brain-computer-interface (BCI) devices has not been extensively explored. In this study, we proposed a high-accuracy MI EEG decoder by incorporating spatial-attention mechanism into convolution neural network (CNN), and deployed it on fully integrated single-chip microcontroller unit (MCU). After the CNN model was trained on workstation computer using GigaDB MI datasets (52 subjects), its parameters were then extracted and converted to build deep-learning architecture interpreter on MCU. For comparison, EEG-Inception model was also trained using the same dataset, and was deployed on MCU. The results indicate that our deep-learning model can independently decode imaginary left-/right-hand motions. The mean accuracy of the proposed compact CNN reaches 96.75 ± 2.41% (8 channels: Frontocentral3 (FC3), FC4, Central1 (C1), C2, Central-Parietal1 (CP1), CP2, C3, and C4), versus 76.96 ± 19.08% of EEG-Inception (6 channels: FC3, FC4, C1, C2, CP1, and CP2). To the best of our knowledge, this is the first portable deep-learning decoder for MI EEG signals. The findings demonstrate high-accuracy deep-learning decoding of MI EEG in a portable mode, which has great implications for hand-disabled patients. Our portable system can be used for developing artificial-intelligent wearable BCI devices, as it is less computationally expensive and convenient for real-life application.

Classification of Motor Imagery EEG Signals Based on Data Augmentation and Convolutional Neural Networks

In brain-computer interface (BCI) systems, motor imagery electroencephalography (MI-EEG) signals are commonly used to detect participant intent. Many factors, including low signal-to-noise ratios and few high-quality samples, make MI classification difficult. In order for BCI systems to function, MI-EEG signals must be studied. In pattern recognition and other fields, deep learning approaches have recently been successfully applied. In contrast, few effective deep learning algorithms have been applied to BCI systems, especially MI-based systems. In this paper, we address these problems from two aspects based on the characteristics of EEG signals: first, we proposed a combined time-frequency domain data enhancement method. This method guarantees that the size of the training data is effectively increased while maintaining the intrinsic composition of the data. Second, our design consists of a parallel CNN that takes both raw EEG images and images transformed through continuous wavelet transform (CWT) as inputs. We conducted classification experiments on a public data set to verify the effectiveness of the algorithm. According to experimental results based on the BCI Competition IV Dataset2a, the average classification accuracy is 97.61%. A comparison of the proposed algorithm with other algorithms shows that it performs better in classification. The algorithm can be used to improve the classification performance of MI-based BCIs and BCI systems created for people with disabilities.

Stockwell transform and semi-supervised feature selection from deep features for classification of BCI signals

Over the past few years, the processing of motor imagery (MI) electroencephalography (EEG) signals has been attracted for developing brain-computer interface (BCI) applications, since feature extraction and classification of these signals are extremely difficult due to the inherent complexity and tendency to artifact properties of them. The BCI systems can provide a direct interaction pathway/channel between the brain and a peripheral device, hence the MI EEG-based BCI systems seem crucial to control external devices for patients suffering from motor disabilities. The current study presents a semi-supervised model based on three-stage feature extraction and machine learning algorithms for MI EEG signal classification in order to improve the classification accuracy with smaller number of deep features for distinguishing right- and left-hand MI tasks. Stockwell transform is employed at the first phase of the proposed feature extraction method to generate two-dimensional time-frequency maps (TFMs) from one-dimensional EEG signals. Next, the convolutional neural network (CNN) is applied to find deep feature sets from TFMs. Then, the semi-supervised discriminant analysis (SDA) is utilized to minimize the number of descriptors. Finally, the performance of five classifiers, including support vector machine, discriminant analysis, k-nearest neighbor, decision tree, random forest, and the fusion of them are compared. The hyperparameters of SDA and mentioned classifiers are optimized by Bayesian optimization to maximize the accuracy. The presented model is validated using BCI competition II dataset III and BCI competition IV dataset 2b. The performance metrics of the proposed method indicate its efficiency for classifying MI EEG signals.

Evaluation of Machine Learning Algorithms for Classification of EEG Signals

In brain–computer interfaces (BCIs), it is crucial to process brain signals to improve the accuracy of the classification of motor movements. Machine learning (ML) algorithms such as artificial neural networks (ANNs), linear discriminant analysis (LDA), decision tree (D.T.), K-nearest neighbor (KNN), naive Bayes (N.B.), and support vector machine (SVM) have made significant progress in classification issues. This paper aims to present a signal processing analysis of electroencephalographic (EEG) signals among different feature extraction techniques to train selected classification algorithms to classify signals related to motor movements. The motor movements considered are related to the left hand, right hand, both fists, feet, and relaxation, making this a multiclass problem. In this study, nine ML algorithms were trained with a dataset created by the feature extraction of EEG signals.The EEG signals of 30 Physionet subjects were used to create a dataset related to movement. We used electrodes C3, C1, CZ, C2, and C4 according to the standard 10-10 placement. Then, we extracted the epochs of the EEG signals and applied tone, amplitude levels, and statistical techniques to obtain the set of features. LabVIEW™2015 version custom applications were used for reading the EEG signals; for channel selection, noise filtering, band selection, and feature extraction operations; and for creating the dataset. MATLAB 2021a was used for training, testing, and evaluating the performance metrics of the ML algorithms. In this study, the model of Medium-ANN achieved the best performance, with an AUC average of 0.9998, Cohen’s Kappa coefficient of 0.9552, a Matthews correlation coefficient of 0.9819, and a loss of 0.0147. These findings suggest the applicability of our approach to different scenarios, such as implementing robotic prostheses, where the use of superficial features is an acceptable option when resources are limited, as in embedded systems or edge computing devices.

Multi-Hierarchical Fusion to Capture the Latent Invariance for Calibration-Free Brain-Computer Interfaces

Brain-computer interfaces (BCI) based motor imagery (MI) has become a research hotspot for establishing a flexible communication channel for patients with apoplexy or degenerative pathologies. Accurate decoding of motor imagery electroencephalography (MI-EEG) signals, while essential for effective BCI systems, is still challenging due to the significant noise inherent in the EEG signals and the lack of informative correlation between the signals and brain activities. The application of deep learning for EEG feature representation has been rarely investigated, nevertheless bringing improvements to the performance of motor imagery classification. This paper proposes a deep learning decoding method based on multi-hierarchical representation fusion (MHRF) on MI-EEG. It consists of a concurrent framework constructed of bidirectional LSTM (Bi-LSTM) and convolutional neural network (CNN) to fully capture the contextual correlations of MI-EEG and the spectral feature. Also, the stacked sparse autoencoder (SSAE) is employed to concentrate these two domain features into a high-level representation for cross-session and subject training guidance. The experimental analysis demonstrated the efficacy and practicality of the proposed approach using a public dataset from BCI competition IV and a private one collected by our MI task. The proposed approach can serve as a robust and competitive method to improve inter-session and inter-subject transferability, adding anticipation and prospective thoughts to the practical implementation of a calibration-free BCI system.

Arithmetic Optimization with RetinaNet Model for Motor Imagery Classification on Brain Computer Interface

Brain Computer Interface (BCI) technology commonly used to enable communication for the person with movement disability. It allows the person to communicate and control assistive robots by the use of electroencephalogram (EEG) or other brain signals. Though several approaches have been available in the literature for learning EEG signal feature, the deep learning (DL) models need to further explore for generating novel representation of EEG features and accomplish enhanced outcomes for MI classification. With this motivation, this study designs an arithmetic optimization with RetinaNet based deep learning model for MI classification (AORNDL-MIC) technique on BCIs. The proposed AORNDL-MIC technique initially exploits Multiscale Principal Component Analysis (MSPCA) approach for the EEG signal denoising and Continuous Wavelet Transform (CWT) is exploited for the transformation of 1D-EEG signal into 2D time-frequency amplitude representation, which enables to utilize the DL model via transfer learning approach. In addition, the DL based RetinaNet is applied for extracting of feature vectors from the EEG signal which are then classified with the help of ID3 classifier. In order to optimize the classification efficiency of the AORNDL-MIC technique, arithmetical optimization algorithm (AOA) is employed for hyperparameter tuning of the RetinaNet. The experimental analysis of the AORNDL-MIC algorithm on the benchmark data sets reported its promising performance over the recent state of art methodologies.

Exploiting pretrained CNN models for the development of an EEG-based robust BCI framework

Identifying motor and mental imagery electroencephalography (EEG) signals is imperative to realizing automated, robust brain-computer interface (BCI) systems. In the present study, we proposed a pretrained convolutional neural network (CNN)-based new automated framework feasible for robust BCI systems with small and ample samples of motor and mental imagery EEG training data. The framework is explored by investigating the implications of different limiting factors, such as learning rates and optimizers, processed versus unprocessed scalograms, and features derived from untuned pretrained models in small, medium, and large pretrained CNN models. The experiments were performed on three public datasets obtained from BCI Competition III. The datasets were denoised with multiscale principal component analysis, and time-frequency scalograms were obtained by employing a continuous wavelet transform. The scalograms were fed into several variants of ten pretrained models for feature extraction and identification of different EEG tasks. The experimental results showed that ShuffleNet yielded the maximum average classification accuracy of 99.52% using an RMSProp optimizer with a learning rate of 0.000 1. It was observed that low learning rates converge to more optimal performances compared to high learning rates. Moreover, noisy scalograms and features extracted from untuned networks resulted in slightly lower performance than denoised scalograms and tuned networks, respectively. The overall results suggest that pretrained models are robust when identifying EEG signals because of their ability to preserve the time-frequency structure of EEG signals and promising classification outcomes.

Deep and Wide Transfer Learning with Kernel Matching for Pooling Data from Electroencephalography and Psychological Questionnaires

Motor imagery (MI) promotes motor learning and encourages brain–computer interface systems that entail electroencephalogram (EEG) decoding. However, a long period of training is required to master brain rhythms’ self-regulation, resulting in users with MI inefficiency. We introduce a parameter-based approach of cross-subject transfer-learning to improve the performances of poor-performing individuals in MI-based BCI systems, pooling data from labeled EEG measurements and psychological questionnaires via kernel-embedding. To this end, a Deep and Wide neural network for MI classification is implemented to pre-train the network from the source domain. Then, the parameter layers are transferred to initialize the target network within a fine-tuning procedure to recompute the Multilayer Perceptron-based accuracy. To perform data-fusion combining categorical features with the real-valued features, we implement stepwise kernel-matching via Gaussian-embedding. Finally, the paired source–target sets are selected for evaluation purposes according to the inefficiency-based clustering by subjects to consider their influence on BCI motor skills, exploring two choosing strategies of the best-performing subjects (source space): single-subject and multiple-subjects. Validation results achieved for discriminant MI tasks demonstrate that the introduced Deep and Wide neural network presents competitive performance of accuracy even after the inclusion of questionnaire data.

Abdul Majeed A. The classification of EEG-based winking signals: a transfer learning and random forest pipeline

Brain Computer-Interface (BCI) technology plays a considerable role in the control of rehabilitation or peripheral devices for stroke patients. This is particularly due to their inability to control such devices from their inherent physical limitations after such an attack. More often than not, the control of such devices exploits electroencephalogram (EEG) signals. Nonetheless, it is worth noting that the extraction of the features and the classification of the signals is non-trivial for a successful BCI system. The use of Transfer Learning (TL) has been demonstrated to be a powerful tool in the extraction of essential features. However, the employment of such a method towards BCI applications, particularly in regard to EEG signals, are somewhat limited. The present study aims to evaluate the effectiveness of different TL models in extracting features for the classification of wink-based EEG signals. The extracted features are classified by means of fine-tuned Random Forest (RF) classifier. The raw EEG signals are transformed into a scalogram image via Continuous Wavelet Transform (CWT) before it was fed into the TL models, namely InceptionV3, Inception ResNetV2, Xception and MobileNet. The dataset was divided into training, validation, and test datasets, respectively, via a stratified ratio of 60:20:20. The hyperparameters of the RF models were optimised through the grid search approach, in which the five-fold cross-validation technique was adopted. The optimised RF classifier performance was compared with the conventional TL-based CNN classifier performance. It was demonstrated from the study that the best TL model identified is the Inception ResNetV2 along with an optimised RF pipeline, as it was able to yield a classification accuracy of 100% on both the training and validation dataset. Therefore, it could be established from the study that a comparable classification efficacy is attainable via the Inception ResNetV2 with an optimised RF pipeline. It is envisaged that the implementation of the proposed architecture to a BCI system would potentially facilitate post-stroke patients to lead a better life quality.

Intelligent Diagnosis towards Hydraulic Axial Piston Pump Using a Novel Integrated CNN Model

As a critical part of a hydraulic transmission system, a hydraulic axial piston pump plays an indispensable role in many significant industrial fields. Owing to the practical undesirable working environment and hidden faults, it is challenging to precisely and effectively detect and diagnose the varying fault in the engineering. Deep learning-based technology presents special strengths in processing mechanical big data. It can simultaneously complete the feature extraction and classification, and achieve the automatic information learning. The popular convolutional neural network (CNN) is exploited for its potent ability of image processing. In this paper, a novel combined intelligent method is developed for fault diagnosis towards a hydraulic axial piston pump. First, the conversion of signals to images is conducted via continuous wavelet transform; the effective feature is preliminarily extracted from the transformed time-frequency images. Second, a novel deep CNN model is constructed to achieve the fault classification. To disclose the potential learning in the disparate layers of the CNN model, the visualization of reduced features is performed by employing t-distributed stochastic neighbor embedding. The effectiveness and stability of the proposed model are validated through the experiments. With the proposed method, different fault types can be precisely identified and high classification accuracy is achieved in a hydraulic axial piston pump.