Multi-channel Physiological Signals Research Articles

Machine-Learning-Based-Approaches for Sleep Stage Classification Utilising a Combination of Physiological Signals: A Systematic Review

Increasingly prevalent sleep disorders worldwide significantly affect the well-being of individuals. Sleep disorder can be detected by dividing sleep into different stages. Hence, the accurate classification of sleep stages is crucial for detecting sleep disorders. The use of machine learning techniques on physiological signals has shown promising results in the automatic classification of sleep stages. The integration of information from multichannel physiological signals has shown to further enhance the accuracy of such classification. Existing literature reviews focus on studies utilising a single channel of EEG signals for sleep stage classification. However, other review studies focus on models developed for sleep stage classification, utilising either a single channel of physiological signals or a combination of various physiological signals. This review focuses on the classification of sleep stages through the integration of combined multichannel physiological signals and machine learning methods. We conducted a comprehensive review spanning from the year 2000 to 2023, aiming to provide a thorough and up-to-date resource for researchers in the field. We analysed approximately 38 papers investigating sleep stage classification employing various machine learning techniques integrated with combined signals. In this study, we describe the models proposed in the existing literature for sleep stage classification, discuss their limitations, and identify potential areas for future research.

A novel convolution bi-directional gated recurrent unit neural network for emotion recognition in multichannel electroencephalogram signals.

Recognising emotions in humans is a great challenge in the present era and has several applications under affective computing. Deep learning (DL) is found as a successful tool for prediction of human emotions in different modalities. To predict 3D emotions with high accuracy in multichannel physiological signals, i.e. electroencephalogram (EEG). A hybrid DL model consisting of convolutional neural network (CNN) and gated recurrent units (GRU) is proposed in this work for emotion recognition in EEG data. CNN has the capability of learning abstract representation, whereas GRU can explore temporal correlation. A bi-directional variation of GRU is used here to learn features in both directions. Discrete and dimensional emotion indices are recognised in two publicly available datasets SEED and DREAMER, respectively. A fused feature of energy and Shannon entropy (𝐸𝑛𝑆𝐸→) and energy and differential entropy (𝐸𝑛𝐷𝐸→) are fed in the proposed classifier to improve the efficiency of the model. The performance of the presented model is measured in terms of average accuracy, which is obtained as 86.9% and 93.9% for SEED and DREAMER datasets, respectively. The proposed convolution bi-directional gated recurrent unit neural network (CNN-BiGRU) model outperforms most of the state-of-the-art and competitive hybrid DL models, which indicates the effectiveness of emotion recognition using EEG signals and provides a scientific base for the implementation in human-computer interaction (HCI).

Respiratory rate estimation from multi-channel signals using auto-regulated adaptive extended Kalman filter

Background:Respiration rate (RR) is a major cause for false alarms in intensive care units (ICU) and is primarily impaired by the artifact prone signals from skin-attached electrodes. Catheter-integrated esophageal electrodes are an alternative source for multi-channel physiological signals from multiple organs such as the heart and the diaphragm. Nonlinear estimation and sensor fusion are promising techniques for extracting the respiratory activity from such multi-component signals, however, pathologic breathing patterns with rapid RR changes typically observed in patient populations such as premature infants, pose significant challenges. Methods:We developed an auto-regulated adaptive extended Kalman filter (AA-EKF), which iteratively adapts the system model and the noise parameters based on the respiratory pattern. AA-EKF was tested on neonatal esophageal observations (NEO), and also on simulated multi-components signals created using waveforms in CapnoBase and ETNA databases. Results:AA-EKF derived RR (RRAA-EKF) from NEO had lower median (inter-quartile range) error of 0.1 (10.6) breaths per minute (bpm) compared to contemporary neonatal ICU monitors (RRNICU): −3.8 (15.7) bpm (p < 0.001). RRAA-EKF error of −0.2 (3.2) bpm was achieved for ETNA wave forms and a bias (95% LOA) of 0.1 (−5.6, 5.9) in breath count. Mean absolute error (MAE) of RRAA-EKF with Capnobase waveforms, as median (inter-quartile range), at 0.3 (0.2) bpm was comparable to the literature reported values. Discussion:The auto-regulated approach allows RR estimation on a broad set of clinical data without requiring extensive patient specific adjustments. Causality and fast response times of EKF based algorithms makes the AA-EKF suitable for bedside monitoring in the ICU setting.

A Continuous Late-Onset Sepsis Prediction Algorithm for Preterm Infants Using Multi-Channel Physiological Signals From a Patient Monitor.

The aim of this study is to develop an explainable late-onset sepsis (LOS) prediction algorithm using continuous multi-channel physiological signals that can be applied to a patient monitor for preterm infants in a neonatal intensive care unit (NICU). The algorithm uses features on heart rate variability (HRV), respiration, and motion, based on electrocardiogram (ECG) and chest impedance (CI). In this study, 127 preterm infants were included, of whom 59 were bloodculture-proven LOS patients and 68 were control patients. Features in 24 hours before the onset of sepsis (LOS group), and an age-matched onset time point (control group) were extracted and fed into machine learning classifiers with gestational age and birth weight. We compared the prediction performance of several well-known classifiers using features from different signal channels (HRV, respiration, and motion) individually as well as their combinations. The prediction performance was evaluated using the area under the receiver-operating-characteristics curve (AUC). The best performance was achieved by an extreme gradient boosting classifier combining features from all signal channels, with an AUC of 0.88, a positive predictive value of 0.80, and a negative predictive value of 0.83 during the 6 hours preceding LOS onset. This feasibility study demonstrates the complementary predictive value of motion information in addition to cardiorespiratory information for LOS prediction. Furthermore, visualization of how each feature in the individual patient impacts the algorithm decision strengthen its interpretability. In clinical practice, it is important to motivate clinical interventions and this visualization method can help to support the clinical decision.

Modality encoded latent dataset for emotion recognition

Variational autoencoder (VAE) is an unsupervised learning that represents high dimensional input data into normally distributed latent space. Multi-channel physiological signals, namely EEG and peripherals are mostly preferred for affective computing. The DEAP dataset is converted into multimodal latent dataset for emotion recognition in this study. 40-ch recordings of 32 participants are encoded to different modalities of peripherals and 32-ch EEG. First, short-time Fourier transform (STFT) is used to extract time–frequency (TF) distribution for training VAE. Thus, the localized components in the each channel of the modalities is converted to 100-dimensional space using VAE. The proposed method is applied to each participant’s recordings to obtain new latent encoded dataset. Within and between subject classification results using latent dataset are compared to the original data for peripheral, 32ch EEG and peripheral with EEG modalities. Naive Bayes (NB) classifier is used to evaluate the encoding performance of the 100-dimensional modalities, and compared to original results. The error rates of leave-one participant-out cross-validation (LOPO CV) 0.3322 and 0.3327 are yielded for high/low arousal and valence states while the originals are 0.349 and 0.382.

A Bayesian Framework for Multivariate Multifractal Analysis

Multifractal analysis has become a reference tool for signal and image processing. Grounded in the quantification of local regularity fluctuations, it has proven useful in an increasing range of applications, yet so far involving only univariate data (scalar valued time series or single channel images). Recently the theoretical ground for multivariate multifractal analysis has been devised, showing potential for quantifying transient higher-order dependence beyond linear correlation among collections of data. However, the accurate estimation of the parameters associated with a multivariate multifractal model remains challenging, especially for small sample size data. This work studies an original Bayesian framework for multivariate multifractal estimation, combining a novel and generic multivariate statistical model, a Whittle-based likelihood approximation and a data augmentation strategy allowing parameter separability. This careful design enables efficient estimation procedures to be constructed for two relevant choices of priors using a Gibbs sampling strategy. Monte Carlo simulations, conducted on synthetic multivariate signals and images with various sample sizes and multifractal parameter settings, demonstrate significant performance improvements over the state of the art, at only moderately larger computational cost. Moreover, we show the relevance of the proposed framework for real-world data modeling in the important application of drowsiness detection from multichannel physiological signals.

FB Dictionary Based SSBL-EM and Its Application for Multi-Class SSVEP Classification Using Eight-Channel EEG Signals

Many applications of signal processing require efficient reconstruction and economical processing of a signal. A classical approach for efficient reconstruction and economical processing of signal is compression. In recent studies, compressed sensing (CS) is an emerging field for compression and reconstruction. Moreover, CS is well-known for energy-efficient signal compression. However, most CS algorithms do not provide efficient reconstruction of physiological signals, such as electroencephalogram (EEG) and electrocardiogram (ECG). Efficient reconstruction is not possible due to non-stationary and non-sparsity characteristics of these signals. On the other hand, processing of these multichannel physiological signals is also time-consuming since the processing time is directly proportional to the number of channels. Hence, we have proposed a Fourier–Bessel (FB) dictionary-based spatiotemporal sparse Bayesian learning (SSBL) algorithm with expectation–maximization (EM) method for efficient reconstruction of multichannel physiological signals. This proposed method is based on the existing SSBL-EM method in which the authors used discrete cosine transform (DCT) dictionary for efficient reconstruction. In our work, we have used FB dictionary instead of DCT dictionary as the Bessel functions have non-stationary characteristic and provide good compression for signals such as EEG. The proposed method has been tested on two different databases for steady-state visual evoked potential (SSVEP)-based brain–computer interface (BCI) classification using multichannel EEG signals and provides 100% classification accuracy with 50% compression and 2-s time duration using the first database. Classification has been performed using L1-regularized multiway canonical correlation analysis (L1-RMCCA) method. The performance of the proposed method has been compared with the DCT-dictionary-based SSBL-EM method. The classification accuracy result makes this proposed method more beneficial for uninterrupted wireless telemonitoring of multichannel EEG signals.

Emotion classification with multichannel physiological signals using hybrid feature and adaptive decision fusion

An effective classification of emotions promotes efficient human-computer and human-human interaction. The diversity of emotion expression and different modal characteristics of multichannel physiological signals render emotion classification based on multichannel physiological signals challenging. This paper presents a multimodal emotion classification framework that uses multichannel physiological signals and introduces two key techniques – hybrid feature extraction and adaptive decision fusion. Firstly, a hybrid feature extraction method is proposed to extract statistical-related and event-related features from physiological signals. The extracted features capture the signal’s statistical and context-dependent characteristics. An adaptive decision fusion method is then proposed to integrate signal modalities for emotion classification. Three different weights are designed by analyzing the physiological signals’ modal and cross-modal information. Finally, the presented framework is evaluated on the Wearable Stress and Affect Detection dataset through a comparative analysis. The results demonstrate the necessity of using event-related features and highlight the importance of developing adaptive decision fusion strategies for emotion classification.

MBioTracker: Multimodal Self-Aware Bio-Monitoring Wearable System for Online Workload Detection.

Cognitive workload affects operators' performance principally in high-risk or time-demanding situations and when multitasking is required. An online cognitive workload monitoring system can provide valuable inputs to decision-making instances, such as the operator's state of mind and resulting performance. Therefore, it can allow potential adaptive support to the operator. This work presents a new design of a wearable embedded system for online cognitive workload monitoring. This new wearable system consists of, on the hardware side, a multi-channel physiological signals acquisition (respiration cycles, heart rate, skin temperature, and pulse waveform) and a low-power processing platform. Further, on the software side, our wearable embedded system includes a novel energy-aware bio-signal processing algorithm. We also use the concept of application self-awareness to enable energy-scalable embedded machine learning algorithms and methods for online subjects' cognitive workload monitoring. Our results show that this new wearable system can continuously monitor multiple bio-signals, compute their key features, and provide reliable detection of high and low cognitive workload levels with a time resolution of 1 minute and a battery lifetime of 14.58 h in our experimental conditions. It achieves a detection accuracy of 76.6% (2.6% lower than analogous offline computer-based analysis) with a sensitivity of 77.04% and a specificity of 81.75%, on a simulated drone rescue mission task. Moreover, by applying our self-aware monitoring to exploit different energy-scalable modes, we can increase battery lifetime by 51.6% (up to 22.11 hours) while incurring an insignificant accuracy loss of 1.07%.

Single-Channel Impedance Plethysmography Neck Patch Device for Unobtrusive Wearable Cardiovascular Monitoring

Background: Wearable and unobtrusive sensing devices are rapidly evolving for long-term cardiovascular monitoring. However, most of the cardiovascular device requires multi-channel physiological signals acquisition, especially in continuous blood pressure (BP) measurement using pulse transition time (PTT) based methods. The multi-devices implementation could impede wearable applications. Objective: This study developed a wearable neck patch device using single-channel impedance plethysmography (IPG) sensing for cardiovascular monitoring, including continuous BP and heart rate (HR) measurement. Methods: IPG-based BP model was derived based on the Bramwell-Hill equation. A patch IPG device was designed and installed above the carotid artery of the subject neck. To validate the BP and HR functions of our device, the Bland-Altman plots were performed to evaluate the estimation error between the reference and the proposed devices within 20 healthy subjects. Results: The BP performance indicates that systolic BP (SBP) estimation error was -0.16 ± 2.97 mmHg and 2.43 ± 1.71 mmHg in terms of mean error (ME) and mean absolute error (MAE), and 0.09 ± 3.30 mmHg and 2.83 ± 1.68 mmHg for diastolic BP (DBP) estimation. Moreover, the HR accuracy has the ME and MAE of 0.02 ± 0.17 bpm and 0.14 ± 0.08 bpm; mean percentage error (MPE) and mean absolute percentage error (MAPE) obtained 0.04 ± 0.23 % and 0.19 ± 0.12 %. Based on statistical results, the BP and HR function of our device satisfied with AAMI/ANSI criteria below 5 ± 8 mmHg and ± 5 bpm or ± 10%. Conclusion: This study implemented a wearable neck patch device with single-channel IPG acquisition that provided two significant cardiovascular parameters of continuous BP and HR, and its performance agreed with standard criteria based on validation with reference sensors. Significance: The proposed proof-of-concept IPG neck patch device has a high potential for wearable applications and low-cost manufacturing in cardiovascular monitoring.

Sleep stage classification using covariance features of multi-channel physiological signals on Riemannian manifolds

Background and objectiveThe recognition of many sleep related pathologies highly relies on an accurate classification of sleep stages. Clinically, sleep stages are usually labelled by sleep experts through visually inspecting the whole-night polysomnography (PSG) recording of patients, wherein electroencephalogram (EEG), electrooculogram (EOG) and electromyogram (EMG) play the dominant role. Developing an automatic sleep staging system based on multi-channel physiological signals could relieve the burden of manual labeling by experts, and obtain reliable and repeatable recognition results as well. MethodsIn this work, we find the correlation between the spatial covariance matrices of multi-channel signals and their corresponding sleep stages. Based on that, we propose two novel sleep stage classification methods based on the features extracted from the covariance matrices of multi-channel signals. Sleep stages are classified using a minimum distance classifier according to their corresponding covariance matrices mapped on Riemannian manifolds. An alternative way to classify these covariance matrices is to represent the features of covariance matrices on the tangent space of Riemannian manifolds and classify them with an ensemble learning classifier. After any of these classification methods, a rule-free refinement process is utilized to further optimize the classification results. ResultsOn the MASS dataset that includes 61 whole-night PSG recordings, both two methods provide satisfactory classification results while the one based on tangent space projection has better performance. On average, an accuracy of 0.812 and a Cohen's Kappa coefficient of 0.722 are obtained under leave-one-subject-out cross validation, using EEG, EOG and EMG signals. Meanwhile, the most effective combinations of EEG channels for sleep staging have been found in this work. ConclusionsThe correlation between spatial covariance matrices of multi-channel signals and their corresponding sleep stages have been found. Features based on that are used for sleep stage classification, and experimental results show the superior performance of proposed methods compared to state-of-the-art works. Results of this work are expected to provide a new vision for dealing with multi-channel or multi-modal signal processing tasks in various applications.

Emotion Recognition Based on Multichannel Physiological Signals with Comprehensive Nonlinear Processing.

Multichannel physiological datasets are usually nonlinear and separable in the field of emotion recognition. Many researchers have applied linear or partial nonlinear processing in feature reduction and classification, but these applications did not work well. Therefore, this paper proposed a comprehensive nonlinear method to solve this problem. On the one hand, as traditional feature reduction may cause the loss of significant amounts of feature information, Kernel Principal Component Analysis (KPCA) based on radial basis function (RBF) was introduced to map the data into a high-dimensional space, extract the nonlinear information of the features, and then reduce the dimension. This method can provide many features carrying information about the structure in the physiological dataset. On the other hand, considering its advantages of predictive power and feature selection from a large number of features, Gradient Boosting Decision Tree (GBDT) was used as a nonlinear ensemble classifier to improve the recognition accuracy. The comprehensive nonlinear processing method had a great performance on our physiological dataset. Classification accuracy of four emotions in 29 participants achieved 93.42%.

Emotion Recognition Based on Weighted Fusion Strategy of Multichannel Physiological Signals.

Emotion recognition is an important pattern recognition problem that has inspired researchers for several areas. Various data from humans for emotion recognition have been developed, including visual, audio, and physiological signals data. This paper proposes a decision-level weight fusion strategy for emotion recognition in multichannel physiological signals. Firstly, we selected four kinds of physiological signals, including Electroencephalography (EEG), Electrocardiogram (ECG), Respiration Amplitude (RA), and Galvanic Skin Response (GSR). And various analysis domains have been used in physiological emotion features extraction. Secondly, we adopt feedback strategy for weight definition, according to recognition rate of each emotion of each physiological signal based on Support Vector Machine (SVM) classifier independently. Finally, we introduce weight in decision level by linear fusing weight matrix with classification result of each SVM classifier. The experiments on the MAHNOB-HCI database show the highest accuracy. The results also provide evidence and suggest a way for further developing a more specialized emotion recognition system based on multichannel data using weight fusion strategy.

Emotion Based Music Recommendation System Using Wearable Physiological Sensors

Most of the existing music recommendation systems use collaborative or content based recommendation engines. However, the music choice of a user is not only dependent to the historical preferences or music contents. But also dependent to the mood of that user. This paper proposes an emotion based music recommendation framework that learns the emotion of a user from the signals obtained via wearable physiological sensors. In particular, the emotion of a user is classified by a wearable computing device which is integrated with a galvanic skin response (GSR) and photo plethysmography (PPG) physiological sensors. This emotion information is feed to any collaborative or content based recommendation engine as a supplementary data. Thus, existing recommendation engine performances can be increased using these data. Therefore, in this paper emotion recognition problem is considered as arousal and valence prediction from multi-channel physiological signals. Experimental results are obtained on 32 subjects' GSR and PPG signal data with/out feature fusion using decision tree, random forest, support vector machine and k-nearest neighbors algorithms. The results of comprehensive experiments on real data confirm the accuracy of the proposed emotion classification system that can be integrated to any recommendation engine.

Emotional empathy transition patterns from human brain responses in interactive communication situations

The paper reports our research aiming at utilization of human interactive communication modeling principles in application to a novel interaction paradigm designed for brain–computer/machine-interfacing (BCI/BMI) technologies as well as for socially aware intelligent environments or communication support systems. Automatic procedures for human affective responses or emotional states estimation are still a hot topic of contemporary research. We propose to utilize human brain and bodily physiological responses for affective/emotional as well as communicative interactivity estimation, which potentially could be used in the future for human–machine/environment interaction design. As a test platform for such an intelligent human–machine communication application, an emotional stimuli paradigm was chosen to evaluate brain responses to various affective stimuli in an emotional empathy mode. Videos with moving faces expressing various emotional displays as well as speech stimuli with similarly emotionally articulated sentences are presented to the subjects in order to further analyze different affective responses. From information processing point of view, several challenges with multimodal signal conditioning and stimuli dynamic response extraction in time frequency domain are addressed. Emotions play an important role in human daily life and human-to-human communication. This is why involvement of affective stimuli principles to human–machine communication or machine-mediated communication with utilization of multichannel neurophysiological and periphery physiological signals monitoring techniques, allowing real-time subjective brain responses evaluation, is discussed. We present our preliminary results and discuss potential applications of brain/body affective responses estimation for future interactive/smart environments.

Research of Tent Map Based Chaotic Particle Swarm Optimization Algorithm for Emotion Recognition

For the problem of feature redundancy of emotion recognition based on multi-channel physiological signals and low efficiency of traditional feature reduction algorithms on great sample data, a new chaotic particle swarm optimization algorithm (TM-CPSO) was proposed to solve the problem of emotion feature selection by combining tent map based chaos search mechanism and improved particle swarm optimization algorithm. The problem of falling into local minimum can be avoided by mapping the search process to the recursive procedure of the chaotic orbit. The recognition rate and efficiency was increased and the algorithm's validity was verified through the analysis of experimental simulation data and the comparison of several recognition methods.