Biomedical Signal Processing

Objective: Studies investigating the complexity in electroencephalography (EEG) in various neuropsychiatric disorders have yielded abnormal results. However, few studies have examined EEG complexity in obsessive-compulsive disorder (OCD). Methods: An eyes-closed scalp EEG series of 3 minutes was recorded in drug-naive patients with OCD and in healthy controls. Each single trial was segmented into multiple identical epochs using two windows of 10 and 30 seconds. Both Kolmogorov Complexity (KC) values and autoregressive (AR) model orders were estimated to quantify the EEG complexity for segmented EEG epochs. Results: The EEG complexity, measured by both KC and AR model orders and in estimations using window lengths of 10 and 30 seconds, was lower in the patients than in the controls. In the AR model orders, the 10-second window differentiated the patients and controls better than the 30-second window. Conclusion: OCD is characterized by low EEG complexity, increased regularity, or decreased randomness. Segmentation of EEG signals is useful for their quantitative identification, a smaller window providing a more sensitive characterization of EEG.

Objective: Some studies on schizophrenia showed an increased complexity in electroencephalography (EEG) whereas others detected a decreased complexity. Because this discrepancy might be due to the clinical features or complexity measures used, we employed two different complexity measures in a group of schizophrenics similar in illness duration (chronic) and symptom profile (residual). Methods: Right-handed chronic residual schizophrenic patients (10 male, 10 female) and age-and sex-matched 20 healthy controls were included in the study. Eyesclosed resting EEG series were measured through quantitative EEG band activities, the log energy entropy (LEE) values, and the Hurst exponents (HE) of EEG measurements were computed for each electrode site. Results: Significantly higher LEE values in the prefrontal, frontal, temporal and parietal locations were observed in schizophrenic patients compared with controls. HE values were significantly higher on the right frontal area in the schizophrenics. Patient group showed increased prefrontal, frontal and parietal delta activity, prefrontal, left temporal and right parietal theta activity and increased left temporal alpha activity. Discussion: In the present study, we found that chronic residual schizophrenia is associated with decreased complexity and increased smoothness in EEG. In addition, EEG of patients was characterized by obvious slowness at prefrontal and frontotemporal regions, dominantly. An integration of EEG complexity and frequency analysis can be proposed as an innovative tool in schizophrenia research.

Texture analysis is a fundamental approach in image processing for identifying specific patterns or structures. One widely used method is the grey-level co-occurrence matrix (GLCM), which computes the frequency of pixel value pairs at certain distances and angles. This study adapts the GLCM method for 1D electroencephalogram (EEG) signal analysis, focusing on extracting features such as contrast, energy, homogeneity, correlation, and entropy. EEG signals are normalized to the range 0-255, and the extracted features are classified using a support vector machine (SVM). Experimental results show that combining features across multiple distances (d=1 to 20) achieves classification accuracy of 78.8% for five classes (Z/O/N/F/S), 94.0% for three classes (O/F/S), and 94.3% for another three-class group (Z/N/F). The method shows strong performance for short to medium distances and fewer class combinations. However, performance declines when dealing with more complex class sets and longer distances, where texture features become less effective. The drop in accuracy for Z/O/N/F/S beyond d=5 underscores the challenges of maintaining feature robustness at extended distances. Despite this, GLCM remains a promising approach for EEG signal classification. Future work should focus on optimizing distance parameters and feature combinations to further enhance classification performance.

With the emergence of quantum computers, traditional cryptographic methods will be compromised and lose their effectiveness. Fortunately, DNA-based cryptographic methods exhibit high resistance against quantum attacks. However, these methods have not yet been employed for securing Persian texts. Additionally, DNA encryption methods require a secure communication channel for sharing the secret key. Therefore, this paper presents a new cryptosystem to ensure the confidentiality of Persian texts based on DNA cryptography. Furthermore, considering the widespread use of images in public media, the capability to encrypt images has also been incorporated into this scheme. In the proposed method, the Hyperelliptic Curve Cryptography (HECC) system is utilized to generate a shared secret key between communicating parties, eliminating the need for a secure channel. DNA hybridization technology is also employed for message encryption and decryption to minimize the computational complexity of the cryptographic algorithm. Performance evaluation results demonstrate that the proposed scheme reduces computational overhead and exhibits lower energy consumption compared to previous approaches.

Traditionally, Blind Speech Separation techniques are computationally expensive as they update the demixing matrix at every time frame index, making them impractical to use in many Real-Time applications. In this paper, a robust data driven twomicrophone sound source localization method is used as a criterion to reduce the computational complexity of the Independent Vector Analysis (IVA) Blind Speech Separation (BSS) method. IVA is used to separate convolutedly mixed speech and noise sources. The practical feasibility of the proposed method is proved by implementing it on a smartphone device to separate speech and noise in Real-World scenarios for Hearing-Aid applications. The experimental results with objective and subjective tests reveal the practical usability of the developed method in many real-world applications.

In this paper, we analyze the performance of electrocardiogram (ECG) signal compression by comparing original and reconstructed signal on two problems. First, automatic sleep stage classification based on ECG signal; second, arrhythmia classification. An effective ECG signal compression method based on two-dimensional wavelet transform which employs set partitioning in hierarchical trees (SPIHT) and beat reordering technique used to compress the ECG signal. This method utilizes the redundancy between adjacent samples and adjacent beats. Beat reordering rearranges beat order in 2D (2 dimension) ECG array based on the similarity between adjacent beats. The experimental results show that the proposed method yields relatively low distortion at high compression rate. The experimental results also show that the accuracy of sleep stage classification and arrhythmia classification using reconstructed ECG signal from proposed method is comparable to the original signal. The proposed method p...

Gastroesophageal reflux disease (GERD) is a common cause of chronic cough. For the diagnosis and treatment of GERD, it is desirable to quantify the temporal correlation between cough and reflux events. Cough episodes can be identified on esophageal manometric recordings as short-duration, rapid pressure rises. The present study aims at facilitating the detection of coughs by proposing an algorithm for the classification of cough events using manometric recordings. The algorithm detects cough episodes based on digital filtering, slope and amplitude analysis, and duration of the event. The algorithm has been tested on in vivo data acquired using a single-channel intra-esophageal manometric probe that comprises a miniature white-light interferometric fiber optic pressure sensor. Experimental results demonstrate the feasibility of using the proposed algorithm for identifying cough episodes based on real-time recordings using a single channel pressure catheter. The presented work can be ...

This study pertains to a six-channel acoustic monitoring system for use in patient monitoring during or after surgery. The base hardware consists of a USB data acquisition system, a custom-built six-channel amplification system, and a series of microphones of various designs. The software is based on the MATLAB platform with data acquisition drivers installed. The displayed information includes: time domain signals, frequency domain signals, and tools to aid in the detection of endobronchial intubation. We hypothesize that the above mentioned arrangement may be helpful to the anesthesiologist in recognizing clinical conditions like wheezing, bronchospasm, endobronchial intubation, and apnea. The study also evaluated various types of microphone designs used to transduce breath sounds. The system also features selectable band-pass filtering using MATLAB algorithms as well as a collection of recordings obtained with the system to establish what respiratory acoustic signals look like un...

Removal of noises from respiratory signal is a classicl problem. In recent years, adaptive filtering has become one of the effective and popular approaches for the processing and analysis of the respiratory and other biomedical signals. Adaptive filters permit to detect time varying potentials and to track the dynamic variations of the signals. Besides, they modify their behavior according to the input signal. Therefore, they can detect shape variations in the ensemble and thus they can obtain a better signal estimation. This paper focuses on (i) Model Respiratory signal with second order Auto Regressive process. Then synthetic noises have been corrupted with respiratory signal and nullify these noises using various adaptive filter algorithms (ii) to remove motion artifacts and 50Hz Power line interference from sinusoidal 0.18Hz respiratory signal using various adaptive filter algorithms. At the end of this paper, a performance study has been done between these algorithms based on various step sizes. It has been found that there will be always tradeoff between step sizes and Mean square error.

We have developed a method that incorporates the time-frequency characteristics of neural sources into magnetoencephalographic (MEG) source estimation. This method, referred to as the time-frequency multiple-signal-classification algorithm, allows the locations of neural sources to be estimated from any time-frequency region of interest. In this paper, we formulate the method based on the most general form of the quadratic time-frequency representations. We then apply it to two kinds of nonstationary MEG data: gamma-band (frequency range between 30-100 Hz) auditory activity data and spontaneous MEG data. Our method successfully detected the gamma-band source slightly medial to the N1m source location. The method was able to selectively localize sources for alpha-rhythm bursts at different locations. It also detected the mu-rhythm source from the alpha-rhythm-dominant MEG data that was measured with the subject's eyes closed. The results of these applications validate the effectiveness of the time-frequency MUSIC algorithm for selectively localizing sources having different time-frequency signatures.

The QRS detection algorithm is substantial for healthcare monitoring and diagnostic applications. A low error detection without adding more computation is a big challenge for researchers. The proposed QRS detection algorithm is a simple, real-time, and high-performance hybrid technique based on decision tree and artificial neural networks (ANN). In this study, the five stages algorithm is designed, implemented, and evaluated for wearable healthcare applications. The first stage is filtering the original ECG signal to reduce the noise and baseline wandering. After that, a maximum or minimum moving-window for positive or negative peaks respectively is searching R-peaks for any expected value and finding the Q and S corresponding to this R-peak. Only these values from all ECG samples are passed to the next stage for feature extraction to reduce the algorithm computation. Stage four is excluded any unlikely points using the mean of the slope and level based on a simple decision tree. Finally, artificial neural networks are designed to classify the rest point for QRS detection using ANNs for each peak polarity to improve the network's performance by separating the data as a positive or negative peak. The algorithm is evaluated based on MATLAB using the MIT-BIH Arrhythmia Database, and the results show a low error rate detection of 0.25%, high sensitivity of 99.86%, and high predictivity of 99.89%. We develop a new approach for real-time QRS detection with low resources and high efficiency compared with other approaches.

The ECG signal processing methods are tested and evaluated based on many databases. The most ECG database used for many researchers is the MIT-BIH arrhythmia database. The QRS-detection algorithms are essential for ECG analyses to detect the beats for the ECG signal. There is no standard number of beats for this database that are used from numerous researches. Different beat numbers are calculated for the researchers depending on the difference in understanding the annotation file. In this paper, the beat numbers for existing methods are studied and compared to find the correct beat number that should be used. We propose a simple function to standardize the beats number for any ECG PhysioNet database to improve the waveform database toolbox (WFDB) for the MATLAB program. This function is based on the annotation's description from the databases and can be added to the Toolbox. The function is removed the non-beats annotation without any errors. The results show a high percentage ...

A technique of analysis of Electroencephalogram signals using wavelet transform and classification using machine learning techniques is developed in this research work. EEG signals are non-stationary that makes the visual analysis time-consuming and may lack quantitative analysis to uncover hidden characteristics of the signals. Artificial Neural Networks along with wavelets provide functions for synthesizing and analyzing the signals and data that exhibits regular behavior punctuated with abrupt changes. This research work concentrates on discrimination between two classes of EEG signals; one acquired from healthy persons and other from epileptic patients. T his article proposes a method for analyzing the brain signals by reliable extraction of attributes using different mother wavelets as Haa r, Coifle t, Daubechies and Sym let and classification of epilepsy disorder using neural networks. An exhaustive analysis points to best suited mother wavelet to extract the attributes that further act as input the machine learning algorithm. The empirical results obtained in this research work indicate that the proposed NN-D ensemble classification method yielded maximum classification accuracy of 99.4% as compared to other ensembles. Hence, it shows the effectiveness of the proposed method for classification of epileptic EEG signals.

This paper presents a Bayesian framework for under-determined audio source separation in multichannel reverberant mixtures. We model the source signals as Student's t latent random variables in a time-frequency domain. The specific structure of musical signals in this domain is exploited by means of a non-negative matrix factorization model. Conversely, we design the mixing model in the time domain. In addition to leading to an exact representation of the convolutive mixing process, this approach allows us to develop simple probabilistic priors for the mixing filters. Indeed, as those filters correspond to room responses they exhibit a simple characteristic structure in the time domain that can be used to guide their estimation. We also rely on the Student's t distribution for modeling the impulse response of the mixing filters. From this model, we develop a variational inference algorithm in order to perform source separation. The experimental evaluation demonstrates the potential of this approach for separating multichannel reverberant mixtures.

Phase recovery of modified spectrograms is a major issue in audio signal processing applications, such as source separation. This paper introduces a novel technique for estimating the phases of components in complex mixtures within onset frames in the Time-Frequency (TF) domain. We propose to exploit the phase repetitions from one onset frame to another. We introduce a reference phase which characterizes a component independently of its activation times. The onset phases of a component are then modeled as the sum of this reference and an offset which is linearly dependent on the frequency. We derive a complex mixture model within onset frames and we provide two algorithms for the estimation of the model phase parameters. The model is estimated on experimental data and this technique is integrated into an audio source separation framework. The results demonstrate that this model is a promising tool for exploiting phase repetitions, and point out its potential for separating overlapping components in complex mixtures.

In this paper we show that considering early contributions of mixing filters through a probabilistic prior can help blind source separation in reverberant recording conditions. By modeling mixing filters as the direct path plus R-1 reflections, we represent the propagation from a source to a mixture channel as an autoregressive process of order R in the frequency domain. This model is used as a prior to derive a Maximum A Posteriori (MAP) estimation of the mixing filters using the Expectation-Maximization (EM) algorithm. Experimental results over reverberant synthetic mixtures and live recordings show that MAP estimation with this prior provides better separation results than a Maximum Likelihood (ML) estimation.

A great number of methods for multichannel audio source separation are based on probabilistic approaches in which the sources are modeled as latent random variables in a Time-Frequency (TF) domain. For reverberant mixtures, it is common to approximate the time-domain convolutive mixing process as being instantaneous in the short-term Fourier transform domain, under a short mixing filters assumption. The TF latent sources are then inferred from the TF mixture observations. In this paper we propose to infer the TF latent sources from the time-domain observations. This approach allows us to exactly model the convolutive mixing process. The inference procedure relies on a variational expectation-maximization algorithm. In significant reverberation conditions, our approach leads to a signal-to-distortion ratio improvement of 5.5 dB compared with the usual TF approximation of the convolutive mixing process.

Abstract:
Cardiovascular diseases are the most common diseases around the world and result in high morbidity and mortality rates. It proves the need to develop new approaches to the disease’s early diagnosis and prevention. Portable health monitoring is characterized by state-of-the-art sensors. It produces a continuous flow of physiological data in real-time. It presents a great opportunity for preventative Cardiovascular disease management. Using deep learning algorithms that is used to accurately and effectively. It creates preventive early diagnosing models for diseases. The authors discuss the integration of wearable health technology and deep learning towards the improvement of individual-oriented technologies in cardiology. The collected data are cleaned for noise and further normalized to provide uniform input in deep learning models. To detect anomalies that are likely to lead to cardiovascular risks, state-of-the-art algorithms. The models are trained and validated using a dataset of wearable device data and clinically diagnosed Cardiovascular disease cases. Wearable health devices combined with deep learning algorithms provide an innovative platform for cardiovascular disease screening and prevention. The incorporation of the proposed system proved its efficiency and accuracy in the identification of potential probable Cardiovascular disease risk, making it possible for early, real-time, noninvasive, and personalized healthcare services. This approach improves early detection but also provides handy information to the patients. This type of research includes the acquisition of larger databases and better models, and incorporating those into more offices and healthcare practices.

This research project aims to explore and compare the effectiveness of Fourier Spectral Analysis (FSA), including Fourier Transform and Spectrogram, alongside Hilbert Huang Transform (HHT) and Hilbert Spectral Analysis (HSA) in analyzing non-linear and nonstationary time series signals.

This research investigates the intricacies of FSA, HHT, and HSA with an in-depth discussion of their mathematical background, working, and limitations with examples. The research applies these methods/frameworks to analyze both synthetic and natural signals and extracts valuable insights by comparing the performance of the two frameworks.

This research also explores the limitations of traditional Fourier Spectral Analysis when dealing with non-stationary and non-linear signals and demonstrates the effectiveness of HHT in providing better time-frequency resolution. The research includes practical applications to real-world sensor data from offshore wind turbines, such as accelerometers, showcasing the potential application of these methods in various scientific and engineering disciplines.

This research also scrutinizes the limitations of HHT and examines recent advancements in the field aimed at addressing these challenges.

We have adopted the original Hérault-Jutten model for the segregation of two concurrent voices in realistic conditions. Firstly, with the new Daimler-Chrysler in-car database, we confirm our previous results obtained with a similar model, and how to make good use of the algorithm. Secondly, we create different mixture conditions from the DC database, and we test 3 kinds of improvement. (1) With the introduction of non-causal FIR filters, the H-J model is able to process asymmetric configurations of two speakers. (2) The benefit of the frequency decomposition is well retrieved with four subbands. Two compatible methods for completing the segregation by post-processing (3) are described: The amplification of the input/output relative level gain by jointenhancement and the beamforming. We show a great overall reduction of the cross-talk, even in noise, which is promising for speech recognition applications.

In the present study, 30 right-handed participants randomly performed one of three motor neurorehabilitation paradigms: action observation (AO), motor imagery (MI) and combined action observation and motor imagery (AO+MI) of the right arm and hand movement. Resting state electroencephalography (EEG) was acquired for 5 minutes before and immediately after the motor paradigms session. EEG was recorded from 10 sites over sensorimotor areas, and the average power was calculated for left (FC3, C3, C1, C5, CP3) and right (FC4, C4, C2, C6, CP4) regions in the spectral bands: delta, theta, alpha, mu, low and high beta. Our main finding demonstrates that delta, theta and mu activity decreased significantly on the contralateral regions during MI, while low beta increased significantly. Except for the mu band, the same changes were observed on the ipsilateral side, where delta and theta decreased significantly, while low beta became significantly higher. No relevant effects were observed for AO or combined AO and MI. These findings demonstrate a rapid effect of MI on cortical modulation in sensorimotor areas which is revealed by changes in resting state oscillatory activity and suggest an interesting interplay between MI and AO. The presented findings may be relevant for choosing a proper protocol for clinical motor neurorehabilitation approaches.

Purpose:The goal of the present study is to quantify the close association between graph theoretic global brain connectivity measures and Alzheimer's Disease (AD) in comparison to Controls. Methods:International Mini-Mental State Examination (MMSE) was used to evaluate cognitive and neuropsychological state of the participants (AD, 12 men, 24 women, mean age = 66.4, sd = 7.9 and controls, 18 men, 11 women, mean age = 67.9, sd = 5.4). There are no comorbidities in patients. Eyesclosed 19-channel surface EEG series were collected from the 2nd Department of Neurology of AHEPA General Hospital of Thessaloniki by experienced neurologists. 2 min long resting-state recordings have been analyzed through non-overlapped sliding window of 1 second and graph theoretical connectivity indices have been estimated by using Directed Transfer Function (DTF) combined with Brain Connectivity Toolbox. EEG recordings and clinical test scores of the individuals were both downloaded from a public dataset on OpenNeuro platform (A dataset of EEG recordings from: Alzheimer's disease, Frontotemporal dementia and Healthy subjects. https:// doi. org/ 10. 18112/ openn euro. ds004 504. v1.0. 7.). Results:AD provided the lower measures in terms of Global Efficiency, Local Efficiency (LE) and Cluster Coefficients. LE estimations provided meaningful and significant statistical difference between patients and controls in theta (4.5-8 Hz), alpha (8.5-12 Hz), beta (12.5-30 Hz), gamma (30.5-45 Hz) sub-bands. Conclusion: The patients provided the lower segregation and integration measures than controls due to loss of connection. AD induces the considerable decrease in segregation. The brain fails to integrate cortical regions into effective networks since there is synaptic disconnection as neuropathology of AD.

In studying neurobiological signals, it has always been a challenge how to gain information from them. It is important to find what is happening in the supposed frequency and time related components of those signals. The results of combined timefrequency domain analysis are very challenging. There are procedures which can give us only time information and those which give only frequency information, but the best methods for non-stationary signals are time-frequency procedures. These can be the most useful procedures for analyzing EEG signals. As we know, the EEG signals are usually nonlinear and non-stationary signals. To analyze them, we need more complicated (or the way to an acceptable solution is complicated) methods then methods based on Fourier transform. The combined procedures of classical methods (up to second order statistical methods, Fourier transform based procedures, correlations, coherence based analysis) and new approaches (DTWDynamical Time Warping, HOSA Higher Orde...

This paper proposes a 2D Non-negative Matrix Factorization (NMF) based single-channel source separation algorithm that emphasizes perceptually important components of audio. Unlike the existing methods, the proposed scheme performs a psychoacoustic pre-processing on the mixture spectrogram in order to supress audio components that are not critical to human hearing sensation while amplifying the perceptually important ones. This yields the auditory spectrogram referred as sonogram of the observed audio mixture and the individual sources are then extracted by 2D NMF. Test results reported in terms of Signal-to-Distortion-Ratio (SDR), Signalto-Inference-Ratio (SIR) and Signal-to-Artifact-Ratio (SAR) show that the proposed perceptually enhanced separation improves the quality of decomposed audio sources by 1.5-6.5 dB with a reduced computational complexity.

Novel minimum-contact vital signs monitoring techniques like textile or capacitive electrocardiogram (ECG) provide new opportunities for health monitoring. These techniques are sensitive to artifacts and require handling of unstable signal quality. Spatio-temporal blind source separation (BSS) is capable of processing suchlike multichannel signals. However, BSS's permutation indeterminacy requires the selection of the cardiac signal (i.e., the component resembling the electric cardiac activity) after its separation from artifacts. This study evaluates different concepts for solving permutation indeterminacy. Methods: Novel automated component selection routines based on heartbeat detections are compared with standard concepts, as using higher order moments or frequencydomain features, for solving permutation indeterminacy in spatio-temporal BSS. BSS was applied to a textile and a capacitive ECG dataset of healthy subjects performing a motion protocol, and to the MIT-BIH Arrhythmia Database. The performance of the subsequent component selection was evaluated by means of the heartbeat detection accuracy (ACC) using an automatically selected single component. Results: The proposed heartbeat-detection-based selection routines significantly outperformed the standard selectors based on Skewness, Kurtosis, and frequency-domain features, especially for datasets containing motion artifacts. For arrhythmia data, beat analysis by sparse coding outperformed simple periodicity tests of the detected heartbeats. Conclusion: Component selection routines based on heartbeat detections are capable of reliably selecting cardiac signals after spatio-temporal BSS in case of severe motion artifacts and arrhythmia. Significance: The availability of robust cardiac component selectors for solving permutation indeterminacy facilitates the usage of spatiotemporal BSS to extract cardiac signals in artifact-sensitive minimum-contact vital signs monitoring techniques.

Novel minimum-contact vital signs monitoring techniques like textile or capacitive electrocardiogram (ECG) provide new opportunities for health monitoring. These techniques are sensitive to artifacts and require handling of unstable signal quality. Spatio-temporal Blind Source Separation (BSS) is capable of processing suchlike multichannel signals. However, BSS's permutation indeterminacy requires the selection of the cardiac signal (i.e. the component resembling the electric cardiac activity) after its separation from artifacts. This study evaluates different concepts for solving permutation indeterminacy. Novel automated component selection routines based on heartbeat detections are compared with standard concepts, as using higher order moments or frequency-domain features, for solving permutation indeterminacy in spatio-temporal BSS. BSS was applied to a textile and a capacitive ECG dataset of healthy subjects performing a motion protocol, and to the MIT-BIH Arrhythmia Databa...

The rapidly rising seizure cases and poor patient-to-neurologist ratio necessitate the development of an efficient automatic seizure detection system. The most commonly used seizure detection systems adopt the patient-dependent approach and lack scalability. Therefore, this work presents an epilepsy seizure detection system using a segment-based patient-independent approach. The designed system is tested on a publicly available Children's Hospital Boston-Massachusetts Institute of Technology Electroencephalogram database. The scheme utilizes a hybrid two-layer feature extraction technique comprising a novel wavelet-based on atomic function, time-domain and nonlinear methods to investigate the EEG signals' ictal and interictal seizure stages. Consequently, the resultant feature set is classified into seizure and non-seizure classes using machine learning algorithms, i.e. K-nearest neighbours, random forest and support vector machine, tuned using the grid search optimization technique. The performance of the proposed system is further improved by the novel ensemble feature selection method, which utilizes the majority voting obtained from different feature selection methods. The distinctiveness of the optimized features is verified by the analysis of variance statistical test. The adaptability and generalizability of the developed patient-independent system is validated on the Neurology and Sleep Centre database. The proposed patient-independent system involving the multi-domain feature extraction and ensemble feature selection provides the highest accuracy of 99.00% using the random forest classifier. Hence, the proposed model proves to be an efficient method for epileptic seizure detection.

This study aims to learn an estimation of the Hurst Parameter for unevenly sampled fractional Brownian motion. The motions are reproduced by means of Cholesky’s algorithm, and the parameter of Hurst is predicted by maximizing likelihoods. These methods are proven to be suitable for the use of quantitative data utilised in this paper during simulation. Several tables that contain the estimates of the self-similarity measure are presented in this study according to various sampling procedures with various sizes of motions. The initiations of these tables is stood on a sequence of statistical tests based on Student’s t-test and Fisher’s Test that make it possible to analyze and compare the distinctions between the considered sampling processes. This paper deals with the simulation of the fractional Brownian motions, with their identifications, and with the analysis of the experimental results. This study proves that unpredictable sampling gives more inconsistency between the outcomes a...

this paper presents a new approach based on the continuous wavelet transform CWT and JADE algorithm for the blind source separation. The JADE algorithm has been widely used to separate the fECG and mECG signals from 8 recordings or channels, while not all of this number is needed. The present work will show that it isn’t the number of channels that counts but rather the quality of the channel used in terms of the energy or the information that it carries. That’s why before introducing the 8 channels to the JADE algorithm we will make a selection by the mathematical microscope Continuous Wavelet Transform CWT and we will show that the number of channels will be reduced to 5 or 3 or even to 2 channels. This algorithm has been validated on several real data. Keywords—blind source separation BSS. continuous wavelet transform. Independent Component Analysis ICA. JADE.

Within a combined EEG-fMRI study of contour integration, we analyze responses to Gabor stimuli with a combined Empirical Mode Decomposition and an Independent Component Analysis. Generaly, responses to different stimuli are very similar thus hard to differentiate. EMD and ICA are used intermingled and not simply in a sequential way. This novel combination helps to suppress redundant modes resulting from an application of ensemble EMD alone. The simulation results show an improved mode separation quality. Hence, the proposed method is an efficient data analysis tool to clearly reveal differences between similar response signals and activity distributions.

Within a combined EEG-fMRI study of contour integration, we analyze responses to Gabor stimuli with an Empirical Mode Decomposition combined with an Independent Component Analysis. Generally, responses to different stimuli are very similar thus hard to differentiate. EMD and ICA are used intermingled and not simply in a sequential way. This novel combination helps to suppress redundant modes resulting from an application of ensemble EMD alone. The simulation results show an improved mode separation quality. Hence, the proposed method is an efficient data analysis tool to clearly reveal differences between similar response signals and activity distributions.

Within a combined EEG-fMRI study of contour integration, we analyze responses to Gabor stimuli with a combined Empirical Mode Decomposition and an Independent Component Analysis. Generaly, responses to different stimuli are very similar thus hard to differentiate. EMD and ICA are used intermingled and not simply in a sequential way. This novel combination helps to suppress redundant modes resulting from an application of ensemble EMD alone. The simulation results show an improved mode separation quality. Hence, the proposed method is an efficient data analysis tool to clearly reveal differences between similar response signals and activity distributions.

Noninvasive medical analyses are a convenient method to study several pathologies even though their indirect nature often requires a complex processing to determine the relevant health "indicators." The usefulness of such indicators depends on the employed model, but also on the uncertainty that is connected to the complex processing involved in the indicator determination. This paper deals with the problems related to the estimation of the uncertainty when the indicators are computed by means of a nontrivial processing on recorded traces of clinical parameters. The paper is focused on the analysis of expired gas traces, but the procedure can also be applied to many other cases where the processing involves manual or automatic selection of suitable "key points" on repetitive traces.

Noninvasive medical analyses are a convenient method to study several pathologies even though their indirect nature often requires a complex processing to determine the relevant health "indicators." The usefulness of such indicators depends on the employed model, but also on the uncertainty that is connected to the complex processing involved in the indicator determination. This paper deals with the problems related to the estimation of the uncertainty when the indicators are computed by means of a nontrivial processing on recorded traces of clinical parameters. The paper is focused on the analysis of expired gas traces, but the procedure can also be applied to many other cases where the processing involves manual or automatic selection of suitable "key points" on repetitive traces.

Suicide, considered as one of the most leading causes of death, has not given enough and appropriate attention in order to reduce its rate such that the humans in all over the world deserve it. The problem addressed in this paper is analysis of the relation between an extra stimulus and physiological data's responses. In order to record the physiological data set from multiple subjects over many weeks, we used an acoustic startle during a Paced Auditory Serial Addition Task (PASAT) test that spontaneously leads subjects to real emotional reactivity, without any deliberate laboratory setting. Crucially, we show that, by inducing anxiety during the test, changes appear in Electrodermal activity, Electrocardiogram, Heart Rate and Respiration Rate. A wide range of physiological features from various analysis domains, including modeling, time/frequency analysis, an algorithm and etc., is proposed in order to find the best emotional reactivity feature to correlate them with emotional states which can be considered as a suicide factor. More specifically, this paper is focused on the EDA data analysis. Experimental results highlight that all cited techniques perform well and we achieved a high resolution of tonic and phasic components which allow us to measure the latency, onsets and amplitudes of EDA responses to a stimulus. This paper follows the association of recommendations for advancement of health care instruments.

This paper relates to the separation of single channel source signals from a single mixed signal by means of independent component analysis (ICA). The proposed idea lies in a time-frequency representation of the mixed signal and the use of ICA on spectral rows corresponding to different time intervals. In our approach, in order to reconstruct true sources, we proposed a novelty idea of grouping statistically independent time-frequency domain (TFD) components of the mixed signal obtained by ICA. The TFD components are grouped by hierarchical clustering and k-mean partitional clustering. The distance between TFD components is measured with the classical Euclidean distance and the β distance of Gaussian distribution introduced by as. In addition, the TFD components are grouped by minimizing the negentropy of reconstructed constituent signals. The proposed method was used to separate source signals from single audio mixes of two- and three-component signals. The separation was performed...

This paper presents a non-neural network approach for the classification of electrocardiogram (ECG) signals. By employing both standard and weighted statistical feature extraction techniques on ECG signals-each comprising 140 data pointsand leveraging dimensionality reduction via Principal Component Analysis (PCA), we build an effective classification pipeline using Support Vector Machines (SVMs). Our method achieves an accuracy of 95.20%, demonstrating that classical machine learning combined with robust feature engineering can rival deep learning approaches in biomedical signal processing.

In this study we propose an automatic method for solving convolutive mixtures separation. The independent components are extracted by frequency domain analysis, where the convolutive model can be solved by instantaneous mixing model approach. The signals are reconstructed back in the observation space resolving the ICA model ambiguities. Simulations are carried out to test the validity of the proposed method

In this paper we compare the performance of different algorithms employed in solving frequency domain blind source separation of convolutive mixtures. The convolutive model is an extension of the instantaneous one and it allows to relax the hypothesis of a linear mixing process in which all the sources are supposed to reach the electrodes at the same time. This test is carried out in the frequency domain, where the algorithms developed for independent component analysis can be employed with minor modifications. The decomposition performance of such algorithms is evaluated on simulated dataset of convultive mixtures of biomedical signals.

Neuroevolution, the amalgamation of neural networks with evolutionary algorithms, stands as a transformative force in advancing Artificial Intelligence (AI). This paper unfolds with the purpose of elucidating the fundamental concepts and applications of Neuroevolution, aiming to provide a nuanced understanding of its significance in propelling the field of AI. Beginning with an exploration of the synergies between evolutionary algorithms and neural networks, the paper emphasizes the overarching objective of showcasing the real-world applicability of Neuroevolution in solving intricate problems across diverse domains. Evolving architectures of neural networks, including the adaptability in Recurrent Neural Networks (RNNs) and Long Short-Term Memory (LSTM) networks, are examined to elucidate the adaptability intrinsic to Neuroevolution. The paper delves into scalability and efficiency strategies, shedding light on handling larger neural network architectures and enhancing computational efficiency. Integration into multi-agent systems is explored, emphasizing Neuroevolution's role in optimizing cooperative and competitive behaviors within complex interactions. Robustness and adaptability analysis of Neuroevolved networks form a critical aspect, evaluating their resilience in varied conditions and their generalization capabilities. Conclusively, the paper outlines the contributions of Neuroevolution to the broader AI landscape, providing insights for researchers and practitioners and fostering developments at the intersection of neural networks and evolutionary algorithms.

Wavelets are functions that satisfy certain mathematical requirement and used in representing data or functions. Wavelets allow complex information such as data compression, signal recognition, signal and image processing, music and computer graphics etc. The wavelet decomposition analysis is used most often in wavelet signal processing. It is used in signal compression as well as in signal identification, although in the latter case, reconstruction of the original is not always required. The decomposition separates a signal into components at various scales corresponding to successive octave frequencies. Each component can be processed individually by a different algorithm. In this work we first try to introduce wavelet and then some of its applications and technique in signal and image processing. Here we also approach a new filtering technique.

Conventional Hidden Markov models generally consist of a Markov chain observed through a linear map corrupted by additive noise. This general class of model has enjoyed a huge and diverse range of applications, for example, speech processing, biomedical signal processing and more recently quantitative finance. However, a lesser known extension of this class of model, is the so-called Factorial Hidden Markov Model (FHMM). FHMMs also have diverse applications, notably in machine learning, artificial intelligence and speech recognition . FHMMs extend the usual class of HMMs, by supposing the partially observed state process is a finite collection of distinct Markov chains, either statistically independent or dependent. There is also considerable current activity in applying collections of partially observed Markov chains to complex action recognition problems, see for example . In this article we consider the Maximum Likelihood (ML) parameter estimation problem for FHMMs. Much of the extant literature concern- * W. P. Malcolm and Novi Quadrianto are with the

This paper presents a new compression scheme for single channel ECG, by delineating each ECG cycle. It uses multirate processing to normalize the varying period beats, followed by amplitude normalization. These beats are coded using vector quantization, with each beat being treated as a vector of uniform dimension. The average amplitude scale factor and the average beat period are made available at the decoder, along with the codebook of period and amplitude normalized beat vectors, to facilitate reconstruction of the signal. The actual beat period and the actual maximum amplitude of the beat are sent to the decoder by DPCM. It achieves a high quality approximation at less than 30 bits per second, with a compression ratio of around 100 : 1 to 200 : 1. To assess the technique properly we have evaluated two measures of error. Finally, the merits and demerits of the technique are discussed.

A method for predicting human muscle performance was developed. Eight test subjects performed a repetitive dynamic exercise to failure using a Lordex spinal machine. Electromyography CEMG) data was collected from the erector spinae. Evaluation ofthe EMG data using a 5 th order Autoregressive CAR) model and statistical regression analysis revealed that an AR parameter, the mean average magnitude of AR poles, can predict performance to failure as early as the second repetition of the exercise. Potential applications to the space program include evaluating on-orbit countermeasure effectiveness, maximizing post-flight recovery, and future real-time monitoring capability during Extravehicular Activity.