Biosignal Analysis Research Articles

MAGNETOENCEPHALOGRAPHY–ELECTROENCEPHALOGRAPHY CO-REGISTRATION USING 3D GENERALIZED HOUGH TRANSFORM

This study proposes an advanced co-registration method for an integrated high temporal resolution electroencephalography (EEG) and magnetoencephalography (MEG) data. The MEG has a higher accuracy for source localization techniques and spatial resolution by sensing magnetic fields generated by the entire brain using multichannel superconducting quantum interference devices, whereas EEG can record electrical activities from larger cortical surface to detect epilepsy. However, by integrating the two modality tools, we can accurately localize the epileptic activity compared to other non-invasive modalities. Integrating the two modality tools is challenging and important. This study proposes a new algorithm using an extended three-dimensional generalized Hough transform (3D GHT) to co-register the two modality data. The pre-process steps require the locations of EEG electrodes, MEG sensors, head-shape points of subjects and fiducial landmarks. The conventional GHT algorithm is a well-known method used for identifying or locating two 2D images. This study proposes a new co-registration method that extends the 2D GHT algorithm to a 3D GHT algorithm that can automatically co-register 3D image data. It is important to study the prospective brain source activity in bio-signal analysis. Furthermore, the study examines the registration accuracy evaluation by calculating the root mean square of the Euclidean distance of MEG–EEG co-registration data. Several experimental results are used to show that the proposed method for co-registering the two modality data is accurate and efficient. The results demonstrate that the proposed method is feasible, sufficiently automatic, and fast for investigating brain source images.

Mindfulness-Based Student Training Leads to a Reduction in Physiological Evaluated Stress.

Background and ObjectiveIn today’s fast-paced modern lifestyle, chronic stress has become a serious issue with potential consequences for our physical and mental health. The concept of mindfulness and its derived Mindfulness-Based Stress Reduction (MBSR) program is considered to be an effective stress management technique for patients as well as for healthy persons. The effects of MBSR interventions on their participants have been subject of previous research, especially with regard to psychological or social science approaches using self-reports and questionnaires. In contrast, medical investigations in this field have been less frequent and often somehow limited, for example, addressing only absolute (discrete) mean values for heart rate or blood pressure.MethodsIn this study, we have evaluated a Mindfulness Based Student Training program (MBST) by applying methods of biosignal analysis to examine its impact on the training participants’ autonomic regulation. This intervention program included classical MBSR elements but was adapted to suit the normal daily needs of university students. We obtained the electrocardiogram, finger-pulse plethysmography, and respiration activity from students participating in either the intervention group (IGR, 38 subjects) or a passive control group (CON, 35 subjects) prior to and after 8 weeks of MBST intervention.ResultsWhen comparing various indices from heart rate variability, pulse wave variability, and respiration in linear and nonlinear domains, significant changes in the autonomic regulation were observed for the IGR group after 8 weeks of MBST.ConclusionThe results indicate a reduced stress level exclusively for the intervention participants, and therefore, we assume a health benefit from the MBST program.

Using a senior Lyapunov exponent to recognize biomedical signals

The article considers the actual problem of analysis of biosignals with chaotic properties. Its solution is important for the recognition of various signals associated with a change in the functional state of the patient during continuous observation. The paper considers the possibility of recognizing atrial fibrillation of rhythmogram signals and anesthesia levels by EEG signals using Lyapunov indicators. The initial data are heart rhythm signals with a duration of 300 cardiocycles and five-second EEG signals. The work shows that the senior Lyapunov indicator allows one to recognize atrial fibrillation against the background of normal rhythm signals and frequent extrasystole, as well as the stage of anesthesia by the EEG signal. The developed algorithm is intended for medical computer systems and is implemented in the MATLAB software environment.

ПРОЄКТУВАННЯ МІЖДИСЦИПЛІНАРНОЇ ІНТЕГРАЦІЇ МЕДИЧНОЇ ІНФОРМАТИКИ

На основі аналізу комплексу наукових дисциплін, які вивчаються студентами спеціальності «Медицина», а саме фундаментальних біомедичних і хімічних, клінічних та соціально-гігієнічних дисциплін, здійснено їх класифікацію для побудови змішаного графу з метою встановлення міждисциплінарної інтеграції з медичною інформатикою. Показано, що для того, щоб сформувати в студентів-медиків цілісне уявлення про професійні сфери застосування інформатики та мотивувати їх до отримання нових знань, необхідно посилити міждисциплінарну інтеграцію дисципліни «Медична інформатика». Результати аналізу зв’язків між елементами субграфів свідчать, що оптимальним є той випадок, коли деякий елемент з кожного кластеру «дисципліна» з’єднаний ребрами з деяким елементом кластера «навчальна тема» з урахуванням множини загальних ознак кожної дисципліни, при цьому відповідно реалізуються випереджальні й зворотні зв'язки змісту навчання. За такої побудови послідовностей зв’язків практичних занять з комплексом наукових дисциплін можна отримати позитивний ефект від застосування міждисциплінарного підходу, оскільки відбуватиметься актуалізація навчального матеріалу, яка сприятиме систематизації й узагальненню знань, формуванню цілісного уявлення про професійні сфери застосування інформатики, створенню мотивації до використання комп’ютерних технологій в професійній діяльності медика. Також побудовано «радіальну» модель, яка відображає поточний стан міждисциплінарних зв’язків та встановлено, що інтеграція навчальних тем реалізується вибірково. Розглянуто ключові аспекти, котрі забезпечують максимальну міждисциплінарну інтеграцію такого практичного заняття, як «Аналіз біосигналів. Методи опрацювання біосигналів. Візуалізація медико-біологічних даних. Оброблення та аналіз медичних зображень». Шляхом анкетування студентів-медиків щодо визначення характеру інтеграції навчальних тем з медичної інформатики та мотивації до використання комп’ютерних технологій в майбутній професійній діяльності отримано результати, які підтверджують доцільність міждисциплінарної інтеграції.

Stable computation of higher order Charlier moments for signal and image reconstruction

The computation of high-order discrete orthogonal moments is very complex and unstable due to fluctuations in numerical polynomial values. In this paper, we will present new algorithms for the stable computation of high-order discrete orthogonal Charlier polynomials and moments. These algorithms are based on the use of modified recurrence relations of Charlier polynomials with respect to the order n and with respect to the variable x, and on the use of the modified Gram-Schmidt orthonormalization process (GSOP). The algorithms in question allow the cancellation of terms that cause the numerical fluctuations of Charlier polynomial values during the recursive calculations. They also preserve the orthogonality property of high-order Charlier polynomials, which enables an efficient analysis of large-size bio-signals and 2D/3D images. Therefore, the simulation results prove the efficiency of the proposed algorithms when it comes to reconstructing large-size bio-signals and images.

Discovering hidden information in biosignals from patients using artificial intelligence.

Biosignals such as electrocardiogram or photoplethysmogram are widely used for determining and monitoring the medical condition of patients. It was recently discovered that more information could be gathered from biosignals by applying artificial intelligence (AI). At present, one of the most impactful advancements in AI is deep learning. Deep learning-based models can extract important features from raw data without feature engineering by humans, provided the amount of data is sufficient. This AI-enabled feature presents opportunities to obtain latent information that may be used as a digital biomarker for detecting or predicting a clinical outcome or event without further invasive evaluation. However, the black box model of deep learning is difficult to understand for clinicians familiar with a conventional method of analysis of biosignals. A basic knowledge of AI and machine learning is required for the clinicians to properly interpret the extracted information and to adopt it in clinical practice. This review covers the basics of AI and machine learning, and the feasibility of their application to real-life situations by clinicians in the near future.

A Wrapper Feature Selection Algorithm: An Emotional Assessment Using Physiological Recordings from Wearable Sensors.

Assessing emotional state is an emerging application field boosting research activities on the topic of analysis of non-invasive biosignals to find effective markers to accurately determine the emotional state in real-time. Nowadays using wearable sensors, electrocardiogram and thoracic impedance measurements can be recorded, facilitating analyzing cardiac and respiratory functions directly and autonomic nervous system function indirectly. Such analysis allows distinguishing between different emotional states: neutral, sadness, and disgust. This work was specifically focused on the proposal of a k-fold approach for selecting features while training the classifier that reduces the loss of generalization. The performance of the proposed algorithm used as the selection criterion was compared to the commonly used standard error function. The proposed k-fold approach outperforms the conventional method with 4% hit success rate improvement, reaching an accuracy near to 78%. Moreover, the proposed selection criterion method allows the classifier to produce the best performance using a lower number of features at lower computational cost. A reduced number of features reduces the risk of overfitting while a lower computational cost contributes to implementing real-time systems using wearable electronics.

Computational analysis of cortical EEG biosignals and neural dynamics underlying an integrated mind-body relaxation technique

Abstract Recent advances in brain computer interfaces (BCI) have brought paradigm shifts in pre-clinical neuroscience research relating various levels of information abstraction in brain. Recent years have seen several neuro electrophysiology studies on traditional and cultural stress reduction methods especially, related to short duration practices that can alleviate workplace stress. With stress reduction using sustainable methods in context, this paper reports on the analysis of biosignals for understanding the neural dynamics in human subjects practicing traditional relaxation techniques using electroencephalography (EEG) technique. The study was conducted among 20 healthy volunteering undergraduate students. Students were randomized as experimental group (N=10), provided yoga and meditation training for 8 weeks, and control group (N=10) with no experience on yoga or meditation. As a reference for analyzing the effect of relaxation techniques in daily life activities, three cognitive tasks, namely, word memory task, psychomotor vigilance task, and steady hand-game task were performed by participants as a pre-EEG recording task. EEG recordings were taken for 160 seconds in eyes-opened relaxed state in a dimly-lit sound-proof environment. In experimental group, after practicing integrated yoga-meditation technique for 8 weeks, a post EEG recording was taken. Relating several functional areas of the brain, cortical activation of the α, β, γ and θ rhythms in the frontal, temporal and parietal lobes of the brain with significant power spectral density variations in experiment group indicating diversity in brain patterns among week-1 to week-8 practitioners was observed. This short-term study indicates the plausibility of integrated yoga-meditation as sustainable relaxation techniques that can influence in improving the coordination in daily life tasks.

Computational Exploration of Neural Dynamics Underlying Music Cues Among Trained and Amateur Subjects

Abstract Exploring the neural basis of music perception and processing has become relevant in neuroscience research for studying human cognition and emotion. Recently many neuroimaging studies have seen brain structure differences in responses to various auditory cue, including music stimuli among diverse individuals. Research still lacks to correlate cortical dynamics in accordance with music familiarity among professionally trained musicians and amateur subjects. With this context, this short-term study reports on effectively recording and computational analysis of neural signals for understanding the cortical network correlates associated with different auditory cues (Western music, Indian classical music, Guitar music and stressor noise) on trained and amateur subjects. On the basis of cognitive battery score analysis, the study recruited 20 volunteering healthy subjects, and grouped as trained subjects (N=10), having music training and amateur subjects (N=10), having no prior music training. EEG recordings were taken for 7 minutes in an eye closed relaxed state, for different auditory stimuli, in a sound proof and dimly lit Indian laboratory settings. In both amateur and trained subjects, heat maps indicated cortical re-organization of ɵ, α, β, and γ rhythms when exposed to different auditory stimuli at similar time bins. Cortical mapping indicated frontal (F3, F4, F7, F8), temporal (T7, T8) and parietal (P7 and P8) as functional regions for processing an auditory information with improved cognitive, attention, memory and language processing skills among trained and amateur subjects in response to familiar music stimuli. With other biosignals analysis techniques and using electrical engineering principles, the study could be further expanded to explore brain anatomy with music therapy as a tool in clinical, therapeutic and community settings.

Continuous and recurrent pattern dynamic neural networks recognition of electrophysiological signals

In the last few years, recurrent and continuous algorithms have became key factors in the solution of diverse pattern recognition problems. The main goal of this study is to introduce four classes of recurrent and continuous artificial neural networks (ANN) that can be implemented for pattern recognition of electrophysiological signals. Such networks are generally known as dynamic neural networks (DNN). The proposed DNN based pattern recognizer uses biosignals raw data as input. This processing method allows capturing the signal time dynamics, which is considered as an intrinsic characteristic of physiology signals. Therefore, recurrent and differential ANN structures were developed to construct different versions of dynamic automatic pattern recognizer. The first one describes the application of Recurrent Neural Networks (RNN) to enforce the biosignal analysis which evolves over time with a fixed sampling period. Three different DNNs with continuous dynamics are introduced. Differential neural network (DifNN) with the capability of learning the evolution of the signal in continuous time, a time-delay neural network (TDNN) for classification is implemented to consider the time-delayed characteristics of the electrophysiological signals and a complex valued neural network (CVNN) which considered the signals to be classified may be pre-processed with a frequency analysis technique. Two different databases of diverse physiological signals are used in this study to validate the application of dynamic neural networks. A first database considers electromiographic (EMG) signals which are tested using the DifNN, TDNN and CVNN. The second database includes gait in Parkinson's disease database signals which are used in the evaluation procedure of RNN. Two validation methods are used to justify the application of dynamic ANNs as pattern recognizer for the EMG activities and the health level classification of patients suffering from Parkinson's: generalization-regularization and the k-fold cross validation. The accuracy estimation and the confusion matrix evaluation confirm the superiority of the proposed approach compared to classical feed-forward ANN pattern recognizer. The particular case of the RNN is also implemented in a 32-bits micro-controller embedded device.

Qualitative Analysis of Brain and Heart Signals using DWT

Bio-signal processing is a widely carried out by the researchers for better understating of complex biological processes. Instead of studying complete biosignal; a decomposed biosignals gives better and more accurate information about the various dynamics involved in the process. Wavelet is one of the powerful transform which is applicable to time and frequency domain. While dealing with various types of physiological signals it is a tedious task to choose the correct or accurate wavelet for the given biosignal analysis. Electrical brain and heart signal along with peripheral pressure signal for 120 subjects is studied by the authors. Authors have checked various quality metrics decide suitability of various Wavelets for EEG, ECG, PP and PPG signal decomposition and reconstruction. The methodology was applied to normal as well as diseased subjects. Our results based on performance parameters like Mean Square Error, Mean Approximate Error, Signal to Error Ratio, PRMSD shows that orthogonal and biorthogonal wavelets are more suitable for bio-signal decomposition and reconstruction. This shows that selection of wavelet should not always be based on similarity between the mother wavelet and the nature of bio-signal.

An EEG-Based Attentiveness Recognition System Using Hilbert\u2013Huang Transform and Support Vector Machine

PurposeAttentiveness recognition benefits the detection of the mental state and concentration when humans perform specific tasks. Hilbert–Huang transform (HHT) is useful for the analysis of nonlinear or nonstationary bio-signals including brainwaves. In this work, a method is proposed for the characterization of attentiveness levels by using electroencephalogram (EEG) signals and HHT analysis.MethodsSingle-channel EEG signals from the frontal area were acquired from participants at different levels of attentiveness and were decomposed into a set of intrinsic mode functions (IMF) by empirical mode decomposition (EMD). Hilbert transform analysis was applied to each IMF to obtain the marginal frequency spectrum. Then the band powers and spectral entropies (SEs) were selected as the attributes of a support vector machine (SVM) for a two-class classification task.ResultsCompared with the predictive models of approximate entropy (ApEn) and fast Fourier transform (FFT), the results show that the band powers extracted from IMF2 to IMF5 of alpha and beta waves and their SE can best discriminate between attentive and relaxed states with the average classification accuracy of 84.80%.ConclusionIn conclusion, this integrated signal processing method is capable of attentiveness recognition that can offer efficient differentiation and may be used in a clinical setting for the detection of attention deficit.

Short-time fractal analysis of biological autoluminescence.

Biological systems manifest continuous weak autoluminescence, which is present even in the absence of external stimuli. Since this autoluminescence arises from internal metabolic and physiological processes, several works suggested that it could carry information in the time series of the detected photon counts. However, there is little experimental work which would show any difference of this signal from random Poisson noise and some works were prone to artifacts due to lacking or improper reference signals. Here we apply rigorous statistical methods and advanced reference signals to test the hypothesis whether time series of autoluminescence from germinating mung beans display any intrinsic correlations. Utilizing the fractional Brownian bridge that employs short samples of time series in the method kernel, we suggest that the detected autoluminescence signal from mung beans is not totally random, but it seems to involve a process with a negative memory. Our results contribute to the development of the rigorous methodology of signal analysis of photonic biosignals.

Serologic Diagnosis of Taenia Solium Cysticercosis through Linear Unmixing Analysis of Biosignals from ACEK Capacitive Sensing Method.

Cysticercosis is a parasitic infection caused by adult tapeworms, and it constantly plagues the livelihoods of people from subsistence farming communities in developing countries. Diagnosis of Cysticercosis typically requires both central nervous system imaging and serological testing. The most common methods in serological testing are Enzyme-linked Immunosorbent Assay (ELISA) and Enzyme Immuno-electrotransfer Blot (EITB). Both ELISA and EITB methods are excessively time-consuming and labor-intensive. Recent research indicates that a shorter assay time and/or higher sensitivity can be achieved by integrating alternate current electrokinetics (ACEK) with biosensing. However, the raw time-series data is very noisy and the size of the dataset is extremely small, which would bring two potential challenges. On one hand, traditional statistical methods cannot extract features robust enough for high sensitivity as well as high specificity. On the other hand, the small data size limits the usage of automatic feature extractors such as deep neural networks. In this paper, we propose a linear unmixing based approach by exploiting the possibility that the time-series biological signals can be represented as linear combinations of source signals. This paper makes distinctive contributions to the field of bio-signal by introducing the unmixing model from the image processing domain to the time-series domain. Experimental results on the classification of Cysticercosis using 123 samples demonstrate the robustness and superior performance of the linear unmixing method over other conventional classifiers in handling small datasets.

Estimation of Sparse VAR Models with Artificial Neural Networks for the Analysis of Biosignals.

Vector autoregressive models (VAR models) are often used to model and to analyze multivariate time series, especially to provide short-term forecasts. A common method of estimating coefficients of these VAR models is solving the Yule- Walker equations. This work introduces and investigates a method to set up "sparse" VAR models, in order to obtain a comparable prognosis quality with significantly fewer coefficients. For this purpose, an artificial neural network was programmed in Python with TensorFlow. Sparsity arises from the implementation of regularization algorithms.Based on simulated data and an ECG, we show that a comparable prognosis quality can be achieved with significantly fewer coefficients. In addition, sparse VAR models can also be determined if the data would actually lead to an underdetermined system of equations. Thus, sparse VAR models may help to classify short epochs of biosignals, e.g. P-waves or QRS-complexes.

A visual-physiology multimodal system for detecting outlier behavior of participants in a reality TV show

This study proposes an outlier detection system based on the visual-physiology multimodal data system for a Korean reality TV show, “Perfect on Paper.” The goal of this system is to provide the program production team and the master of ceremonies with real-time facial expressions and bio-signal analysis results of program participants. Using this information, the program production team and master of ceremonies can easily identify the participants with unusual behavior. We propose an insole-type hardware that measures the wearer’s skin conductivity, temperature, and motion. We also suggest a dynamic time warp-based clustering algorithm for the outlier and micro-expression detection technique. The system was developed starting from the program planning phase through a 6-month period in collaboration with the production team. The developed system analyzed the biometric information and facial expressions of 88 participants in 11 episodes and provided rapid feedback to the production team at the shooting spot in real time. Finally, we present a case where a visual-physiology multimodal system can be useful for a real TV broadcasting production.

DECODING OF HAND GESTURES BY FRACTAL ANALYSIS OF ELECTROMYOGRAPHY (EMG) SIGNAL

One of the major research areas in analysis of human movements is to investigate how different movements are related to biosignals. Hand gestures belong to major movements of human that have been considered widely by researchers. Therefore, decoding of different hand’s gestures by analysis of related biosignal is very important to be considered. In this paper, we analyze the complex structure of electromyography (EMG) signal from subjects who did eight hand gestures. For this purpose, we chose fractal dimension as the indicator of complexity. The analysis showed that the EMG signal has the greatest and lowest fractal dimensions in case of fingers flexed together in fist, and pointing index, respectively. The employed method in this research is not limited to the analysis of the influence of hand’s gestures on EMG signal. However, it can be widely applied to analyze the influence of different types of stimuli on different human’s biosignals.

FRACTAL-BASED CLASSIFICATION OF ELECTROMYOGRAPHY (EMG) SIGNAL IN RESPONSE TO BASIC MOVEMENTS OF THE FINGERS

Investigating human movements is the most important issue in rehabilitation science. Movements of fingers as one of the major movements of human has been considered by many scientists. Therefore, decoding of finger movements by analysis of related biosignal is very important to consider. In this research, we do the complexity analysis on the Electromyography (EMG) signal that was recorded due to basic movements of fingers. In fact, the EMG signal was classified in case of different movements of fingers by fractal analysis. The result of analysis showed that the EMG signal has the greatest and lowest fractal dimension (complexity) in case of thumb finger flexion and little finger extension. In further attempts, the fractal theory can be applied to investigate the influence of other types of stimulation on variations of the complexity of muscles’ reactions.

Analysis of the EEG bio-signals during the reading task by [formula omitted] method

The process of reading a specific text is considered complex and little known in neuroscience, since it involves the vision, memory, motor control, learning, among others. In this sense, an excellent possibility to study the brain activity in the reading task can be achieved by the analysis of the multi-channel Electroencephalogram (EEG) and also with new statistical methods, like the detrended fluctuation analysis method (DFA). In this paper it will be proposed a model to analyze the brain activity in the reading task, performed by two subjects using a 22-channels EEG (NEUROMAP® model EQSA260). In order to test our model, two adults subjects (graduates) were tested here. These subjects were arranged in a chair facing a panel with the specific text, excluding involuntary movements that activated regions of the brain that were not being stimulated by reading. For the first subject, chosen at random, the text was presented before the task for understanding and some memorization. For the other subject the text was presented at the time of task. For the signal processing we chose 11 bio-electrodes located at the frontal, parietal, temporal and occipital regions of the brain. Therefore, to treat these non-stationary bio-signals we must apply robust and modern statistical techniques. With this objective, DFA method was applied in order to analyze the FDFA(n) fluctuation function in multi-channel EEG bio-sensors, more specifically the difference of its logarithm, i.e., ΔlogFDFA. The results show that the use of this new function can be useful for brain activities. This paper, as we shall see here, is an initial contribution for EEG data analyze, that would be of medical interest, mainly in neuroscience area.

Використання фракталів для аналізу біосигналів

Використання фракталів для аналізу біосигналів