Continuous Wavelet Transform for Decoding Finger Movements From Single-Channel EEG

Gamma oscillation optimally predicts finger movements

Our fingers are the most dexterous and complicated parts of our body and play a significant role in our daily activities. Non-invasive techniques, such as Electroencephalography (EEG) and Electromyography (EMG) can be used to collect neural and muscular signals related to finger movements. In this study, we combined an 8-channel EMG and a 31-channel EEG while the human subject moved one of the five fingers on the right hand. To identify the best EEG frequency features that encode distinct finger movements, we systematically examined the decoding accuracies of the slow-cortical potentials and three types of sensorimotor rhythms, namely the Mu, beta, and gamma oscillations. For both EMG and EEG, we came up with a simple and unified root mean square or power approach that avoided the complex signal features used by previous studies. The signal features were then fed into a feedforward artificial-neural-network (ANN) classifier. We found that the low-gamma oscillation provided the best decoding performance over the other frequency bands, ranging from 65.0% to 89.0%, which was comparable to the EMG performance. Combining EMG and low gamma into a single ANN can further improve the outcome for subjects who had showed suboptimal performances with EMG or EEG alone. This study provided a simple and efficient algorithm for prosthetics that assist patients with sensorimotor impairments.

Dynamical Embedding of Single-Channel Electroencephalogram for Artifact Subspace Reconstruction

This study introduces a novel framework to apply the artifact subspace reconstruction (ASR) algorithm on single-channel electroencephalogram (EEG) data. ASR is known for its ability to remove artifacts like eye-blinks and movement but traditionally relies on multiple channels. Embedded ASR (E-ASR) addresses this by incorporating a dynamical embedding approach. In this method, an embedded matrix is created from single-channel EEG data using delay vectors, followed by ASR application and reconstruction of the cleaned signal. Data from four subjects with eyes open were collected using Fp1 and Fp2 electrodes via the CameraEEG android app. The E-ASR algorithm was evaluated using metrics like relative root mean square error (RRMSE), correlation coefficient (CC), and average power ratio. The number of eye-blinks with and without the E-ASR approach was also estimated. E-ASR achieved an RRMSE of 43.87% and had a CC of 0.91 on semi-simulated data and effectively reduced artifacts in real EEG data, with eye-blink counts validated against ground truth video data. This framework shows potential for smartphone-based EEG applications in natural environments with minimal electrodes.

DSFE: Decoding EEG-Based Finger Motor Imagery Using Feature-Dependent Frequency, Feature Fusion and Ensemble Learning

Accurate decoding finger motor imagery is essential for fine motor control using EEG signals. However, decoding finger motor imagery is particularly challenging compared with ordinary motor imagery. This paper proposed a novel EEG decoding method of feature-dependent frequency band selection, feature fusion, and ensemble learning (DSFE) for finger motor imagery. First, a feature-dependent frequency band selection method based on correlation coefficient (FDCC) was proposed to select feature-specific effective bands. Second, a feature fusion method was proposed to fuse different types of candidate features to produce multiple refined sets of decoding features. Finally, an ensemble model using the weighted voting strategy was proposed to make full use of these diverse sets of final features. The results on a public EEG dataset of five fingers motor imagery showed that the DSFE method is effective and achieves the highest decoding accuracy of 50.64%, which is 7.64% higher than existing studies using exactly the same data. The experiments further revealed that both the effective frequency bands of different subjects and the effective frequency bands of different types of features are different in finger motor imagery. Furthermore, compared with two-hand motor imagery, the effective decoding information of finger motor imagery is transferred to the lower frequency. The idea and findings in this paper provide a valuable perspective for understanding fine motor imagery in-depth.

An optimized EEGNet decoder for decoding motor image of four class fingers flexion

As a cutting-edge technology of connecting biological brain and external devices, brain-computer interface (BCI) exhibits promising applications on extensive fields such as medical and military. As for the disable individuals with four limbs losing the motor functions, it is a potential treatment way to drive mechanical equipments by the means of non-invasive BCI, which is badly depended on the accuracy of the decoded electroencephalogram (EEG) singles. In this study, an explanatory convolutional neural network namely EEGNet based on SimAM attention module was proposed to enhance the accuracy of decoding the EEG singles of index and thumb fingers for both left and right hand using sensory motor rhythm (SMR). An average classification accuracy of 72.91% the data of eight healthy subjects was obtained, which were captured from the one second before finger movement to two seconds after action. Furthermore, the character of event-related desynchronization (ERD) and event related synchronization (ERS) of index and thumb fingers was also studied in this study. These findings have significant importance for controlling external devices or other rehabilitation equipment using BCI in a fine way.

Exploring the potential of pretrained CNNs and time-frequency methods for accurate epileptic EEG classification: a comparative study

Prompt diagnosis of epilepsy relies on accurate classification of automated electroencephalogram (EEG) signals. Several approaches have been developed to characterize epileptic EEG data; however, none of them have exploited time-frequency data to evaluate the effect of tweaking parameters in pretrained frameworks for EEG data classification. This study compares the performance of several pretrained convolutional neural networks (CNNs) namely, AlexNet, GoogLeNet, MobileNetV2, ResNet-18 and SqueezeNet for the localization of epilepsy EEG data using various time-frequency data representation algorithms. Continuous wavelet transform (CWT), empirical Fourier decomposition (EFD), empirical mode decomposition (EMD), empirical wavelet transform (EWT), and variational mode decomposition (VMD) were exploited for the acquisition of 2D scalograms from 1D data. The research evaluates the effect of multiple factors, including noisy versus denoised scalograms, different optimizers, learning rates, single versus dual channels, model size, and computational time consumption. The benchmark Bern-Barcelona EEG dataset is used for testing purpose. Results obtained show that the combination of MobileNetV2, Continuous Wavelet Transform (CWT) and Adam optimizer at a learning rate of 10-4, coupled with dual-data channels, provides the best performance metrics. Specifically, these parameters result in optimal sensitivity, specificity, f1-score, and classification accuracy, with respective values of 96.06%, 96.15%, 96.08%, and 96.10%. To further corroborate the efficacy of opted pretrained models on exploited Signal Decomposition (SD) algorithms, the classifiers are also being simulated on Temple University database at pinnacle modeling composition. A similar pattern in the outcome readily validate the findings of our study and robustness of deep learning models on epilepsy EEG scalograms.The conclusions drawn emphasize the potential of pretrained CNN-based models to create a robust, automated system for diagnosing epileptiform. Furthermore, the study offers insights into the effectiveness of varying time-frequency techniques and classifier parameters for classifying epileptic EEG data.

A fused multi-subfrequency bands and CBAM SSVEP-BCI classification method based on convolutional neural network

For the brain-computer interface (BCI) system based on steady-state visual evoked potential (SSVEP), it is difficult to obtain satisfactory classification performance for short-time window SSVEP signals by traditional methods. In this paper, a fused multi-subfrequency bands and convolutional block attention module (CBAM) classification method based on convolutional neural network (CBAM-CNN) is proposed for discerning SSVEP-BCI tasks. This method extracts multi-subfrequency bands SSVEP signals as the initial input of the network model, and then carries out feature fusion on all feature inputs. In addition, CBAM is embedded in both parts of the initial input and feature fusion for adaptive feature refinement. To verify the effectiveness of the proposed method, this study uses the datasets of Inner Mongolia University of Technology (IMUT) and Tsinghua University (THU) to evaluate the performance of the proposed method. The experimental results show that the highest accuracy of CBAM-CNN reaches 0.9813 percentage point (pp). Within 0.1-2 s time window, the accuracy of CBAM-CNN is 0.0201-0.5388 (pp) higher than that of CNN, CCA-CWT-SVM, CCA-SVM, CCA-GNB, FBCCA, and CCA. Especially in the short-time window range of 0.1-1 s, the performance advantage of CBAM-CNN is more significant. The maximum information transmission rate (ITR) of CBAM-CNN is 503.87 bit/min, which is 227.53 bit/min-503.41 bit/min higher than the above six EEG decoding methods. The study further results show that CBAM-CNN has potential application value in SSVEP decoding.

Characteristics of glucose change in diabetes mellitus generalized through continuous wavelet transform processing: A preliminary study

BACKGROUND

The continuous glucose monitoring (CGM) system has become a popular evaluation tool for glucose fluctuation, providing a detailed description of glucose change patterns. We hypothesized that glucose fluctuations may contain specific information on differences in glucose change between type 1 diabetes mellitus (T1DM) and type 2 diabetes mellitus (T2DM), despite similarities in change patterns, because of different etiologies. Unlike Fourier transform, continuous wavelet transform (CWT) is able to simultaneously analyze the time and fre-quency domains of oscillating data.

AIM

To investigate whether CWT can detect glucose fluctuations in T1DM.

METHODS

The 60-d and 296-d glucose fluctuation data of patients with T1DM (n = 5) and T2DM (n = 25) were evaluated respectively. Glucose data obtained every 15 min for 356 d were analyzed. Data were assessed by CWT with Morlet form (n = 7) as the mother wavelet. This methodology was employed to search for limited frequency glucose fluctuation in the daily glucose change. The frequency and enclosed area (0.02625 scalogram value) of 18 emerged signals were compared. The specificity for T1DM was evaluated through multiple regression analysis using items that demonstrated significant differences between them as explanatory variables.

RESULTS

The high frequency at midnight (median: 75 Hz, cycle time: 19 min) and middle frequency at noon (median: 45.5 Hz, cycle time: 32 min) were higher in T1DM vs T2DM (median: 73 and 44 Hz; P = 0.006 and 0.005, respectively). The area of the > 100 Hz zone at midnight to forenoon was more frequent and larger in T1DM vs T2DM. In a day, the lower frequency zone (15-35 Hz) was more frequent and the area was larger in T2DM than in T1DM. The three-dimensional scatter diagrams, which consist of the time of day, frequency, and area of each signal after CWT, revealed that high frequency signals belonging to T1DM at midnight had a loose distribution of wave cycles that were 17-24 min. Multivariate analysis revealed that the high frequency signal at midnight could characterize T1DM (odds ratio: 1.33, 95% confidence interval: 1.08-1.62; P = 0.006).

CONCLUSION

CWT might be a novel tool for differentiate glucose fluctuation of each type of diabetes mellitus using CGM data.

Fine-Grained and Multiple Classification for Alzheimer's Disease With Wavelet Convolution Unit Network

In this article, we propose a novel wavelet convolution unit for the image-oriented neural network to integrate wavelet analysis with a vanilla convolution operator to extract deep abstract features more efficiently. On one hand, in order to acquire non-local receptive fields and avoid information loss, we define a new convolution operation by composing a traditional convolution function and approximate and detailed representations after single-scale wavelet decomposition of source images. On the other hand, multi-scale wavelet decomposition is introduced to obtain more comprehensive multi-scale feature information. Then, we fuse all these cross-scale features to improve the problem of inaccurate localization of singular points. Given the novel wavelet convolution unit, we further design a network based on it for fine-grained Alzheimer's disease classifications (i.e., Alzheimer's disease, Normal controls, early mild cognitive impairment, late mild cognitive impairment). Up to now, only a few methods have studied one or several fine-grained classifications, and even fewer methods can achieve both fine-grained and multi-class classifications. We adopt the novel network and diffuse tensor images to achieve fine-grained classifications, which achieved state-of-the-art accuracy for all eight kinds of fine-grained classifications, up to 97.30%, 95.78%, 95.00%, 94.00%, 97.89%, 95.71%, 95.07%, 93.79%. In order to build a reference standard for Alzheimer's disease classifications, we actually implemented all twelve coarse-grained and fine-grained classifications. The results show that the proposed method achieves solidly high accuracy for them. Its classification ability greatly exceeds any kind of existing Alzheimer's disease classification method.

Real-Time and Automatic System for Performance Evaluation of Karate Skills Using Motion Capture Sensors and Continuous Wavelet Transform

In sports science, the automation of performance analysis and assessment is urgently required to increase the evaluation accuracy and decrease the performance analysis time of a subject. Existing methods of performance analysis and assessment are either performed manually based on human experts’ opinions or using motion analysis software, i.e., biomechanical analysis software, to assess only one side of a subject. Therefore, we propose an automated system for performance analysis and assessment that can be used for any human movement. The performance of any skill can be described by a curve depicting the joint angle over the time required to perform a skill. In this study, we focus on only 14 body joints, and each joint comprises three angles. The proposed system comprises three main stages. In the first stage, data are obtained using motion capture inertial measurement unit sensors from top professional fighters/players while they are performing a certain skill. In the second stage, the collected sensor data obtained are input to the biomechanical software to extract the player’s joint angle curve. Finally, each joint angle curve is processed using a continuous wavelet transform to extract the main curve points (i.e., peaks and valleys). Finally, after extracting the joint curves from several top players, we summarize the players’ curves based on five statistical indicators, i.e., the minimum, maximum, mean, and mean  $\pm$  standard deviation. These five summarized curves are regarded as standard performance curves for the joint angle. When a player’s joint curve is surrounded by the five summarized curves, the performance is considered acceptable. Otherwise, the performance is considered unsatisfactory. The proposed system is evaluated based on four different karate skills. The results of the proposed system are identical to the decisions of the expert panels and are thus suitable for real-time decisions.

Human Gait Activity Recognition Machine Learning Methods

Human gait activity recognition is an emerging field of motion analysis that can be applied in various application domains. One of the most attractive applications includes monitoring of gait disorder patients, tracking their disease progression and the modification/evaluation of drugs. This paper proposes a robust, wearable gait motion data acquisition system that allows either the classification of recorded gait data into desirable activities or the identification of common risk factors, thus enhancing the subject's quality of life. Gait motion information was acquired using accelerometers and gyroscopes mounted on the lower limbs, where the sensors were exposed to inertial forces during gait. Additionally, leg muscle activity was measured using strain gauge sensors. As a matter of fact, we wanted to identify different gait activities within each gait recording by utilizing Machine Learning algorithms. In line with this, various Machine Learning methods were tested and compared to establish the best-performing algorithm for the classification of the recorded gait information. The combination of attention-based convolutional and recurrent neural networks algorithms outperformed the other tested algorithms and was individually tested further on the datasets of five subjects and delivered the following averaged results of classification: 98.9% accuracy, 96.8% precision, 97.8% sensitivity, 99.1% specificity and 97.3% F1-score. Moreover, the algorithm's robustness was also verified with the successful detection of freezing gait episodes in a Parkinson's disease patient. The results of this study indicate a feasible gait event classification method capable of complete algorithm personalization.

ICA With CWT and k-means for Eye-Blink Artifact Removal From Fewer Channel EEG

In recent years, there has been an increase in the usage of consumer based EEG devices with fewer channel configuration. Although independent component analysis has been a popular approach for eye-blink artifact removal from multichannel EEG signals, several studies showed that there is a leak of neural information into the eye-blink artifact associated independent components (ICs). Furthermore, the leak increases as the number of input EEG channels decreases and leads to loss of valuable EEG information. To overcome this problem, we developed a new framework that combines ICA with continuous wavelet transform (CWT), k- means and singular spectrum analysis (SSA) methods. In contrast to the existing approaches, the artifact region in the identified eye-blink artifact IC is detected and suppressed rather than setting it to zero as in classical ICA. As most of the energy in the eye-blink artifact IC is concentrated in the artifact region, CWT and k- means algorithms exploits this feature to detect the eye-blink artifact region. Support vector machine (SVM) based classifier is finally designed for automatic detection of the eye blink artifact ICs. The performance of proposed method is evaluated on synthetic and two real EEG datasets for various EEG channels setting. Results highlight that for fewer channel EEG signals, the proposed method provides accurate separation without any neural information loss as compared to the existing methods.

On the characterization of cognitive tasks using activity-specific short-lived synchronization between electroencephalography channels

Accurate characterization of brain activity during a cognitive task is challenging due to the dynamically changing and the complex nature of the brain. The majority of the proposed approaches assume stationarity in brain activity and disregard the systematic timing organization among brain regions during cognitive tasks. In this study, we propose a novel cognitive activity recognition method that captures the activity-specific timing parameters from training data that elicits maximal average short-lived pairwise synchronization between electroencephalography signals. We evaluated the characterization power of the activity-specific timing parameter triplets in a motor imagery activity recognition framework. The activity-specific timing parameter triplets consist of latency of the maximally synchronized signal segments from activity onset Δt, the time lag between maximally synchronized signal segments τ, and the duration of the maximally synchronized signal segments w. We used cosine-based similarity, wavelet bi-coherence, phase-locking value, phase coherence value, linearized mutual information, and cross-correntropy to calculate the channel synchronizations at the specific timing parameters. Recognition performances as well as statistical analyses on both BCI Competition-III dataset IVa and PhysioNet Motor Movement/Imagery dataset, indicate that the inter-channel short-lived synchronization calculated using activity-specific timing parameter triplets elicit significantly distinct synchronization profiles for different motor imagery tasks and can thus reliably be used for cognitive task recognition purposes.

Decoding Finger Tapping With the Affected Hand in Chronic Stroke Patients During Motor Imagery and Execution

In stroke rehabilitation, motor imagery based on a brain-computer interface is an extremely useful method to control an external device and utilize neurofeedback. Many studies have reported on the classification performance of motor imagery to decode individual fingers in stroke patients compared with healthy controls. However, classification performance for a given limb is still low because the differences between patients owing to brain reorganization after stroke are not considered. We used electroencephalography signals from eleven healthy controls and eleven stroke patients in this study. The subjects performed a finger tapping task during motor execution, and motor imagery was performed with the dominant and affected hands in the healthy controls and stroke patients, respectively. All fingers except for the thumb were classified using the proposed framework based on a voting module. The averaged four-class accuracies during motor execution and motor imagery were 53.16 ± 8.42% and 46.94 ± 5.99% for the healthy controls and 53.17 ± 14.09% and 66.00 ± 14.96% for the stroke patients, respectively. Importantly, the classification accuracies in the stroke patients were statistically higher than those in healthy controls during motor imagery. However, there was no significant difference between the accuracies of motor execution and motor imagery. These findings show the potential for high classification performance for a given limb during motor imagery in stroke patients. These results can also provide insights into controlling an external device on the basis of a brain-computer interface.

Eye-blink artifact removal from single channel EEG with k-means and SSA

In recent years, the usage of portable electroencephalogram (EEG) devices are becoming popular for both clinical and non-clinical applications. In order to provide more comfort to the subject and measure the EEG signals for several hours, these devices usually consists of fewer EEG channels or even with a single EEG channel. However, electrooculogram (EOG) signal, also known as eye-blink artifact, produced by involuntary movement of eyelids, always contaminate the EEG signals. Very few techniques are available to remove these artifacts from single channel EEG and most of these techniques modify the uncontaminated regions of the EEG signal. In this paper, we developed a new framework that combines unsupervised machine learning algorithm (k-means) and singular spectrum analysis (SSA) technique to remove eye blink artifact without modifying actual EEG signal. The novelty of the work lies in the extraction of the eye-blink artifact based on the time-domain features of the EEG signal and the unsupervised machine learning algorithm. The extracted eye-blink artifact is further processed by the SSA method and finally subtracted from the contaminated single channel EEG signal to obtain the corrected EEG signal. Results with synthetic and real EEG signals demonstrate the superiority of the proposed method over the existing methods. Moreover, the frequency based measures [the power spectrum ratio ([Formula: see text]) and the mean absolute error (MAE)] also show that the proposed method does not modify the uncontaminated regions of the EEG signal while removing the eye-blink artifact.

Appropriate Mother Wavelets for Continuous Gait Event Detection Based on Time-Frequency Analysis for Hemiplegic and Healthy Individuals

Gait event detection is a crucial step towards the effective assessment and rehabilitation of motor dysfunctions. Recently, the continuous wavelet transform (CWT) based methods have been increasingly proposed for gait event detection due to their robustness. However, few investigations on determining the appropriate mother wavelet with proper selection criteria have been performed, especially for hemiplegic patients. In this study, the performances of commonly used mother wavelets in detecting gait events were systematically investigated. The acceleration signals from the tibialis anterior muscle of both healthy and hemiplegic subjects were recorded during ground walking and the two core gait events of heel strike (HS) and toe off (TO) were detected from the signal recordings by a CWT algorithm with different mother wavelets. Our results showed that the overall performance of the CWT algorithm in detecting the two gait events was significantly different when using various mother wavelets. By using different wavelet selection criteria, we also found that the accuracy criteria based on time-error minimization and F1-score maximization could provide the appropriate mother wavelet for gait event detection. The findings from this study will provide an insight on the selection of an appropriate mother wavelet for gait event detection and facilitate the development of adequate rehabilitation aids.