Signal Analysis Research Articles

Tutorial. Frequency analysis of the surface EMG signal: Best practices

This tutorial is aimed primarily to non-engineers (clinical researchers, clinicians, neurophysiology technicians, ergonomists, movement and sport scientists, physical therapists) or beginners using, or planning to use, surface electromyography (sEMG) as a monitoring and assessment tool for muscle and neuromuscular evaluations in the prevention and rehabilitation fields.Its first purpose is to explain, with minimal mathematics, basic concepts related to: (a) time and frequency domain description of a signal, (b) Fourier transform, (c) amplitude, phase, and power spectrum of a signal, (d) sampling of a signal, (e) filtering of sEMG signals, (f) cross-spectrum and coherence between two signals, (g) signal stationarity and criteria for epoch selection, (h) myoelectric manifestations of muscle fatigue and (i) fatigue indices. These concepts are consolidated knowledge and are addressed and discussed with examples taken from the literature.

Research on magnetic flux leakage testing of pipelines by finite element simulation combined with artificial neural network

Magnetic flux leakage (MFL) testing technology is widely employed in non-destructive testing of pipelines, and the analysis of leakage signals plays a crucial role in assessing pipelinea safety. This paper introduces a novel approach for MFL testing, which combines finite element simulation with artificial neural networks. First, a finite element model for MFL testing of defects is established, the influence of magnetization states on MFL signals is discussed, and the variation of signal extremum with magnetization intensity is analyzed. Next, suitable MFL signal features are selected to focus on the relationship between defect types, defect sizes, and these features. Finally, a kernel extreme learning machine (KELM) predictive model is developed to classify defect types and predict defect sizes. The results indicate that as magnetization intensity increases, the magnetization process of the pipeline can be divided into a nonlinear growth phase and a linear phase, with MFL signal extremum rapidly increasing and then gradually growing linearly. Different geometric features of defects correspond to distinct distributions of MFL signals, effectively reflecting variations in defect types and sizes. Compared to traditional ELM models, the KELM model achieves higher prediction accuracy and stable performance, with the radial basis kernel function significantly enhancing the generalization and predictive capabilities of the neural network.

Exploring EEG-Based Biomarkers for Improved Early Alzheimer's Disease Detection: A Feature-Based Approach Utilizing Machine Learning

This paper presents a comprehensive investigation into Electroencephalogram (EEG) signal processing and analysis techniques aimed at enhancing early diagnosis methods for Alzheimer's Disease (AD). Leveraging a dataset that has EEG data of individuals diagnosed with Mild Cognitive Impairment (MCI), AD, Healthy Controls, and the study explores Preprocessing Methods and Feature Extraction Techniques, with machine learning model notably Support Vector Machines (SVM).In the preprocessing phase, a combination of high pass, lowpass, Savitzky–Golay, and median filters are applied, informed by a comprehensive review of filter comparison literature. Feature extraction encompasses three primary categories: ‘Statistical, ‘Frequency Domain’ and ‘Time Domain’. The scope of this work is to explore features in all these three domains and build SVM based model for efficient classification. In our investigation, we achieved a categorization accuracy of 92% through the utilization of statistical features. Employing time domain features resulted in an accuracy of 87%, while frequency domain features also yielded an 87% accuracy rate in our study. The primary objective of this study is that it aims to enhance early AD diagnosis through advanced EEG signal processing and machine learning techniques, focusing on preprocessing methods, feature extraction, and classification accuracy.

Multi Modal AI-Driven System for Enhanced Neurological Detection Using EEG, Retinal, and Brain Imaging Data

This study presents a novel approach to detecting neurological disorders by integrating electroencephalogram (EEG) signal analysis with retinal scans, brain imaging techniques (MRI and CT), and patient-reported data. The primary objective is to develop an AI-driven diagnostic tool that utilizes multi-modal data fusion and deep learning for feature extraction, significantly enhancing diagnostic accuracy and supporting personalized patient profiles. Advanced deep learning architectures are employed to identify meaningful patterns from each data source, while innovative fusion techniques synthesize these patterns to deliver comprehensive assessments. Key outcomes demonstrate substantial improvements in diagnostic precision, early detection capabilities, and real-time monitoring, facilitating timely clinical decision support. Validation through extensive testing on diverse datasets confirms a marked increase in diagnostic accuracy. The practical implications of this system include the early and accurate detection of neurodegenerative diseases, epilepsy, and psychiatric disorders, ultimately contributing to improved patient outcomes and tailored healthcare solutions.

STATISTICAL ANALYSIS OF FRICTIONAL PROCESSESOF A SLIDING BEARING

This article addresses the issue of diagnosing friction nodes in sliding bearings, which are among the mostchallenging and complex topics in the tribological research. The article presents a methodology for diagnosingfriction nodes operating under variable loads based on the analysis of vibration acceleration signals from thebearing using statistical methods. In the study, statistical parameters were determined to assess the technicalcondition of sliding bearings operating with variable loads under dry and mixed friction conditions. Usingstatistical tools to interpret selected tribological parameters that occur in the friction process aims to ensure theproper operation of the tribological system under operating conditions. Analysis of the measurement resultspresented in the study using statistical tools will allow the development of a methodology to assess the correctoperation of sliding bearings.

Retraction notice: Machine learning based pervasive analytics for ECG signal analysis

Retraction notice: Machine learning based pervasive analytics for ECG signal analysis

Мирон Пилипович Деркач (1931–2012). Той, що почав розмовляти з комп’ютером…

Myron Derkach is a world-class scientist in the field of biophysics and informational aspects of speech communication – analysis, synthesis and recognition of speech signals. He is the founder of the scientific direction, who formulated the principles of the deductive strategy of auditory language perception. In 1976, he created one of the first biophysics departments in Ukraine on the basis of the biological faculty of Ivan Franko University of Lviv. Dr. Myron Derkach is the author of the first university textbooks in Ukrainian on biophysics and mathematical methods in biology, which have not lost their relevance to this day.

Two dimensional computational model of calcium and IP3 interacting system dynamics in an obese hepatocyte cell

Abstract Calcium ion (Ca$^{2+}$) signaling is crucial in regulating numerous cellular processes vital for preserving structural integrity and functional equilibrium across diverse cell types. Both the calcium stores and mitochondria play significant roles in this signaling pathway. The calcium source may be in the form of a blip or a puff depending on the various conditions of the cellular systems. The one dimensional model of calcium dynamics with IP$_{3}$ gives crucial insight of feedback mechanisms influencing calcium homeostasis. In order to obtain deeper insights of local impacts of various mechanisms and feedbacks in hepatocyte cell, it is necessary to develop the models in higher dimensions. In order to get more deeper insights, two dimensional model is proposed assuming the phenomena to be uniform along z dimension. This research presents a two-dimensional computational model to analyse the interactive system dynamics of inositol 1,4,5-trisphosphate (IP$_{3}$) and Ca$^{2+}$, aiming to assess how these signaling patterns influence hepatocyte functionality which allows to incorporate puff type of calcium source under both obese and normal physiological states. It further examines the implications of calcium signaling on NADH synthesis, ATP production, and degradation rates. Numerical simulations are executed utilising the Crank-Nicolson method for temporal analysis and the Linear Finite Element Method for spatial analysis. Additionally, the study conducts a comparative analysis of calcium signaling between obese and normal hepatocyte. The findings offer enhanced insights into the interactive system dynamics of IP$_{3}$ and Ca$^{2+}$ in hepatocytes, elucidating the effects of various parameter alterations on cellular behaviour in both states.

A genome engineered tool set for Drosophila TGF-β/BMP signaling studies.

Ligands of the TGF-β/BMP superfamily are critically involved in the regulation of growth, patterning and organogenesis and can act as long-range morphogens. Essential for understanding TGF-β/BMP signaling dynamics and regulation are tools that allow monitoring and manipulating pathway components expressed at physiological levels and endogenous spatiotemporal patterns. We used genome engineering to generate a comprehensive library of endogenously epitope- or fluorescently-tagged versions of receptors, co-receptors, transcription factors and key feedback regulators of the Drosophila BMP and Activin signaling pathways. We demonstrate that the generated alleles are biologically active and can be utilized for assessing tissue and subcellular distribution of the corresponding proteins. Further, we show that the genomic platforms can be used for in locus structure-function and cis-regulatory analyses. Finally, we present a complementary set of protein binder-based tools, which allow visualization as well as manipulation of the stability and subcellular localization of epitope-tagged proteins, providing new tools for the analysis of BMP signaling and beyond.

A non-invasive heart rate prediction method using a convolutional approach.

The research focuses on leveraging convolutional neural networks (CNNs) to enhance the analysis of physiological signals, specifically photoplethysmogram (PPG) data which is a valuable tool for non-invasive heart rate prediction. Recognizing the global challenge of heart failure, the study seeks to provide a rapid, accurate, and non-invasive alternative to traditional, uncomfortable blood pressure cuffs. To achieve more accurate and efficient heart rate estimates, a k-fold CNN model with an optimal number of convolutional layers is employed. While the findings show promising results, the study addresses potential sources of error in cuffless PPG-based heart rate measurement, including motion artifacts and skin color variations, emphasizing the need for validation against gold standard methods. The research optimizes a CNN model with optimal layers, operating on 1D arrays of 8-s data slices and employing k-fold cross-validation to mitigate probabilistic uncertainties. Finally, the model yields a remarkable minimum absolute error (MAE) rate of 6.98 beats per minute (bpm), marking a significant 10% improvement over recent studies. The study also advances medical diagnostics and data analysis, then lays a strong foundation for developing cost-effective, reliable devices that offer a more comfortable and efficient way of predicting heart rate.

Location Method of Buried Polyethylene Gas Pipeline Based on Acoustic Signal Ellipse Method

This study proposes a buried PE gas pipeline positioning method based on the elliptical method of an acoustic signal analysis. The cross-correlation time delay positioning technology is combined with the elliptical equation, forming an effective mechanism for pipeline depth positioning. First, a dual-tree complex wavelet transform is employed to denoise the collected signals, enhancing the quality and accuracy of the data. Subsequently, the cross-correlation function is utilized to extract the delay times between the signals. The obtained delay times are then substituted into the elliptical equation to calculate the depth of the buried PE pipeline. Based on this theoretical framework, a simulation model is established in COMSOL, and positioning simulation analyses are conducted under three different conditions: pipeline depth, relative sensor positions, and distances between sensors and excitation points. The simulation results indicate that a clear correlation exists between the signal delay time and the pipeline position, with simulation errors controlled within 5%, thus validating the theoretical feasibility of the method. To further assess the effectiveness of this approach, an experimental testing system is constructed. The experimental study was carried out under four different conditions: pipeline burial depth, relative sensor positions, distances between sensors and excitation points, and excitation frequencies. The experimental results demonstrate that these factors significantly affect the pipeline depth positioning. The comparison results show that the method has a high accuracy in depth positioning, with experimental errors controlled within 10%. This study proves that accurate positioning of pipeline depth could be achieved by substituting signal delay times into the elliptical equation, thereby validating the method’s feasibility in practical applications. The proposed method effectively addressed the shortcomings of existing pipeline depth positioning technologies, providing important theoretical support and a practical reference for future pipeline positioning research.

Improving automated labeling with deep learning and signal segmentation for accurate ECG signal analysis

Improving automated labeling with deep learning and signal segmentation for accurate ECG signal analysis

Current signal analysis using SW-GAT networks for fault diagnosis of electromechanical drive systems under extreme data imbalance

Abstract In complex industrial environments, ensuring the safe operation and effective maintenance of electromechanical equipment is of paramount importance. Intelligent fault diagnosis based on deep learning is currently the most popular data-driven method. However, conventional intelligent fault diagnosis techniques face several challenges: (1) Most diagnostic models rely heavily on analyzing vibration signals. However, vibration sensors are difficult to deploy in space-constrained environments, and vibration signals are frequently contaminated by strong noise. (2) The prevalence of class imbalance between normal and fault data in equipment condition monitoring can lead to model over-reliance on information from a few classes. (3) Traditional diagnostic models presuppose data independence, neglecting the coupling relationships between data. To address the aforementioned issue, this paper proposes a Self-Weighted Graph Attention Networks (SW-GAT) based on motor stator current signal analysis, aimed at solving the fault diagnosis problem of critical transmission components in electromechanical systems under severely imbalanced data scenarios. Firstly, the raw current data is preprocessed using Stacked Auto-Encoders (SAE), and then the decoded current frequency-domain data is utilized to construct graphical data, thereby enhancing the non-common features and weak fault information in the current signals. Secondly, by introducing the graph pooling attention mechanism into GAT, the model can more effectively focus on useful fault feature information within the graph data. Finally, a novel interclass adjustment loss function is designed to adaptively adjust and balance class weights, enabling the model to pay greater attention to minority class samples and thereby improving the recognition accuracy for minority class faults. Validating the proposed method on two cases and comparing it with other advanced approaches, our method achieved the highest accuracy among the compared methods.

Kinematic signature of high risk labored breathing revealed by novel signal analysis.

Breathing patterns (respiratory kinematics) contain vital prognostic information. This dimension of physiology is not captured by conventional vital signs. We sought to determine the feasibility and utility of quantifying respiratory kinematics. Using inertial sensors, we analyzed upper rib, lower rib, and abdominal motion of 108 patients with respiratory symptoms during a hospital encounter (582 two-minute recordings). We extracted 34 features based on an explainable correspondence with well-established breathing patterns. K-means clustering revealed that respiratory kinematics had three dimensions apart from the respiratory rate. We represented these dimensions using respiratory rate variability, respiratory alternans (rib-predominant breaths alternating with abdomen-predominant ones), and recruitment of accessory muscles (increased upper rib excursion). Latent profile analysis of the kinematic measures revealed two profiles consistent with the established clinical constructs of labored and unlabored breathing. In logistic regression, the labored breathing profile improved model discrimination for critical illness beyond the Sequential Organ Failure Assessment (SOFA) score (AUROC 0.77 v/s 0.72; p = 0.02). These findings quantitatively confirm the prior understanding that the respiratory rate alone does not adequately represent the complexity of respiratory kinematics; they demonstrate that high-dimensional signatures of labored breathing can be quantified in routine practice settings, and they can improve predictions of clinical deterioration.

Response reconstruction based on measurement matrix optimization in compressed sensing for structural health monitoring

Structural health monitoring (SHM) data have a large volume, increasing the cost of data storage and transmission and the difficulties of structural parameter identification. The compressed sensing (CS) theory provides a signal acquisition and analysis strategy. Signal reconstruction using limited measurements and CS has attracted significant interest. However, the dynamic responses obtained from civil engineering structures contain noise, resulting in sparse samples and reducing the signal reconstruction accuracy. Therefore, we propose an optimization algorithm for the measurement matrix integrating the Karhunen-Loeve transform (KLT) and approximate QR decomposition (KLT-QR) to improve the accuracy of dynamic response reconstruction of SHM data. The KLT reduces the correlation between the measurement matrix and the sparse basis. The approximate QR decomposition is used to improve the independence between the column vectors of the measurement matrix, optimizing the measurement matrix. The experimental results for a laboratory steel beam indicate that the proposed KLT-QR algorithm outperforms three other algorithms regarding the accuracy of dynamic response reconstruction (acceleration, displacement, and strain), especially at high compression ratios. The acceleration responses from the Ji’an Bridge are utilized to verify the advantages of the proposed algorithm. The results demonstrate that the KLT-QR algorithm has the highest accuracy of reconstructing the vibration signals and yields better Fourier spectra than the conventional Gaussian measurement matrix.

A review on deep learning methods for heart sound signal analysis

IntroductionApplication of Deep Learning (DL) methods is being increasingly appreciated by researchers from the biomedical engineering domain in which heart sound analysis is an important topic of study. Diversity in methodology, results, and complexity causes uncertainties in obtaining a realistic picture of the methodological performance from the reported methods.MethodsThis survey paper provides the results of a broad retrospective study on the recent advances in heart sound analysis using DL methods. Representation of the results is performed according to both methodological and applicative taxonomies. The study method covers a wide span of related keywords using well-known search engines. Implementation of the observed methods along with the related results is pervasively represented and compared.Results and discussionIt is observed that convolutional neural networks and recurrent neural networks are the most commonly used ones for discriminating abnormal heart sounds and localization of heart sounds with 67.97% and 33.33% of the related papers, respectively. The convolutional neural network and the autoencoder network show a perfect accuracy of 100% in the case studies on the classification of abnormal from normal heart sounds. Nevertheless, this superiority against other methods with lower accuracy is not conclusive due to the inconsistency in evaluation.

Relativistic time delay analysis of pulsar signals near ultracompact objects

Relativistic time delay analysis of pulsar signals near ultracompact objects

Slidingan analysis of analytical signal of non-contact photoplethysmography for assessing heart rate

The paper proposes a method for studying the variability of the subject’s heart rate based on the intellectual analysis of the pulse wave measured with remote photoplethysmography. The logically related stages of the formation of quadrature components based on the Hilbert transform of biomedical signals’ dynamics are presented. Within the framework of modern methods of intellectual analysis of non-stationary time series, realizations of adaptive estimates of instantaneous frequencies and periods of the heartbeat basic tone are obtained.

LC dynamic signal modeling and analysis of resonant frequency measurement error under rapid eye movements interference

Abstract LC resonant wireless passive intraocular pressure (IOP) sensors measure IOPs by mapping them to LC resonance frequencies. However, during the sweeping acquisition process of each LC signal frame, possible rapid eye movement (REM) can interfere with the wireless mutual inductance coupling, leading to the signal distortion and resulting in the error of resonance frequency measuring. Currently, there is a lack of modelling analysis of the errors generated by REM effect, resulting in an absence of guidance for reducing the impact. Therefore, here, we start from the perspective of the signal model and establish the LC dynamic signal model for sweeping acquisition. By means of the limit state analysis and Monte Carlo simulation, we analyze the influence of external REM parameters (including REM range and velocity) and internal parameters of the LC sensing system (including the quality coefficient of LC sensor, diameter, and number of turns of the reader coil, signal sweep acquisition speed and period) on the errors. We demonstrated theoretical relationship between the extreme errors with these parameters and verify it through a computer simulation. Based on the results, we propose to optimize the internal parameters of the LC sensing system to reduce the REMs effect on errors, safeguarding the quality of signal acquisition and improving the measurement accuracy.

External-cavity diode laser at 2.6 µm and its frequency stabilization with a scanning Fabry-Pérot cavity

We report on the construction of an external cavity diode laser at 2.6 µm wavelength and its frequency stabilization with an external scanning Fabry-Pérot cavity using Pound-Drever-Hall (PDH) frequency stabilization technique. We discuss the frequency noise characteristics of the laser using the PDH residual error signal analysis and the locking performance. We have achieved a 2.5 x 105 suppression factor in the frequency noise power spectral density at Fourier frequencies lower than 100 Hz when the laser is stabilized. The laser is employed to observe resonant excitation on the 5s5p3P0 →5s4d3D1 transition in cold 88Sr atoms trapped in a magneto-optical trap. We have observed an enhancement in the trapped atom number by a factor of 8.3 compared to the case when a single laser at 707 nm is employed for repumping.