Second-order Blind Identification Research Articles

Identifying Structural Damage in Steel Truss Frames Using Computer-Based Numerical Simulation with the Second Order Blind Identification (SOBI) Method

Identifying Structural Damage in Steel Truss Frames Using Computer-Based Numerical Simulation with the Second Order Blind Identification (SOBI) Method

Self-gated free-running 5D whole-heart MRI using blind source separation for automated cardiac motion extraction.

To compare two blind source separation (BSS) techniques to principal component analysis and the electrocardiogram for the identification of cardiac triggers in self-gated free-running 5D whole-heart MRI. To ascertain the precision and robustness of the techniques, they were compared in three different noise and contrast regimes. The repeated superior-inferior (SI) projections of a 3D radial trajectory were used to extract the physiological signals in three cardiac MRI cohorts: (1) 9 healthy volunteers without contrast agent injection at 1.5T, (2) 30 ferumoxytol-injected congenital heart disease patients at 1.5T, and (3) 12 gadobutrol-injected patients with suspected coronary artery disease at 3T. Self-gated cardiac triggers were extracted with the three algorithms (principal component analysis [PCA], second-order blind identification [SOBI], and independent component analysis [ICA]) and the difference with the electrocardiogram triggers was calculated. PCA and SOBI triggers were retained for image reconstruction. The image sharpness was ascertained on whole-heart 5D images obtained with PCA and SOBI and compared among the three cohorts. SOBI resulted in smaller trigger differences in Cohorts 1 and 3 compared to PCA (p < 0.01) and in all cohorts compared to ICA (p < 0.04). In Cohorts 1 and 3, the sharpness increased significantly in the reconstructed images when using SOBI instead of PCA (p < 0.03), but not in Cohort 2 (p = 0.4). We have shown that SOBI results in more precisely extracted self-gated triggers than PCA and ICA. The validation across three diverse cohorts demonstrates the robustness of the method against acquisition variability.

Operating modal parameter identification of railway vehicle bogie based on k-value optimization variational mode decomposition and second-order blind identification method

Due to the variable operating conditions and strong background noise, the accuracy of modal parameter identification of the railway vehicle bogie is low. This paper proposes a method for identifying operational modal parameters by combining the K-value optimization variational mode decomposition (KVMD) and second-order blind identification (SOBI) method (KVMD–SOBI method). Firstly, the variational mode decomposition (VMD) algorithm is used to adaptively decompose a single-channel vibration signal into intrinsic mode function (IMF) components with different scale characteristics. The feature IMF components are selected through the K-value optimization method based on energy ratio and stability graph. Then, the SOBI algorithm is used to separate the matrix composed of the feature IMF components. And the mode shape and single-mode signal are obtained. Finally, the natural frequency and damping ratio of each order mode are extracted by the single-mode parameter recognition method. Through simulation and experimental verification, it is shown that the proposed method has good noise robustness and can effectively identify the first three modal parameters of railway vehicle bogies in variable-speed operation. At the same time, the feature IMF components extracted by KVMD algorithm eliminate the sensitivity of sensor installation positions and operating conditions to detection results. Therefore, this method is an excellent operational mode identification method based on a single-channel signal with good noise and working condition robustness.

Patterns identification using blind source separation with application to neural activities associated with anticipated falls

Exploring the nature of neural activity in humans attracts much attention in human-machine interfaces. However, there is still a lack of understanding about the neural activities associated with human anticipated falls. This study examines how the brain responds to anticipated falls by identifying the patterns of neural activities associated with anticipated falls with Electroencephalograms (EEG). We developed a method for pattern identification of neural activities with second-order blind identification in the machine learning framework. The effects of artifact removal within EEG data are compared between the second-order blind identification (SOBI) algorithm and a related method. Experiments involving 135 anticipated falls and 135 walk controls show that the median sample entropy of the EEG signals in the anticipated falls is from 0.1151 to 0.2267 smaller (p-value <0.01) than those in the walk controls. In anticipated falls detection, the average accuracy with random forests is 85.2% for channel O2, which is nearly 10% higher than the channel temporal cerebral cortex (T3 and T4). The results can further guide us in developing new systems of human–machine interfaces, such as EEG data-augmented wearable airbags, to prevent anticipated falls.

Multi-Task AI-driven undetermined blind source identification for modal identification of large scale structures

The effective management of aging infrastructure represents a critical concern globally, exerting profound socio-economic influences and occupying a prominent position on the political agenda. The escalating occurrences of catastrophic collapses in recent years underscore the pressing requirement for efficient Structural Health Monitoring (SHM) strategies. These strategies play a pivotal role in aiding decision-making processes related to prioritizing interventions and rehabilitation efforts. Among SHM systems, those utilizing Operational Modal Analysis (OMA) through vibration-based techniques have gained popularity due to their non-destructive nature, comprehensive damage assessment capabilities, and relative ease of automation. Despite their advantages, current OMA methods encounter significant scalability challenges, primarily attributable to extensive computational demands and the necessity for substantial expert involvement. In this context, recent advancements in the field of artificial intelligence (AI) present a promising solution to address these scalability issues, opening the door for the development of next-generation SHM systems. This study introduces an innovative Multitask Learning Deep Neural Network (MTL-DNN) model designed for rapid and automated blind source modal identification of structures. Integrating the concepts of undetermined second-order blind source identification (SOBI) into the architecture of the network, the proposed model can reveal the independent modal components present in the raw response acceleration data with minimum computational burden. The presented results demonstrate the quasi-instantaneous modal identification capabilities of the proposed architecture, enabling the estimation of more components than the number of sensors.

Expanding the clinical application of OPM-MEG using an effective automatic suppression method for the dental brace metal artifact

Optically pumped magnetometer magnetoencephalography (OPM-MEG) holds significant promise for clinical functional brain imaging due to its superior spatiotemporal resolution. However, effectively suppressing metallic artifacts, particularly from devices such as orthodontic braces and vagal nerve stimulators remains a major challenge, hindering the wider clinical application of wearable OPM-MEG devices.A comprehensive analysis of metal artifact characteristics from time, frequency, and time–frequency perspectives was conducted for the first time using an OPM-MEG device in clinical medicine. This study focused on patients with metal orthodontics, examining the modulation of metal artifacts by breath and head movement, the incomplete regular sub-Gaussian distribution, and the high absolute power ratio in the 0.5–8 Hz band. The existing metal artifact suppression algorithms applied to SQUID-MEG, such as fast independent component analysis (FastICA), information maximization (Infomax), and algorithms for multiple unknown signal extraction (AMUSE), exhibit limited efficacy. Consequently, this study introduced the second-order blind identification (SOBI) algorithm, which utilized multiple time delays for the component separation of OPM-MEG measurement signals. We modified the time delays of the SOBI method to improve its efficacy in separating artifact components, particularly those in the ultralow frequency range. This approach employs the frequency-domain absolute power ratio, root mean square (RMS) value, and mutual information methods to automate the artifact component screening process.The effectiveness of this method was validated through simulation experiments involving four subjects in both resting and evoked experiments. In addition, the proposed method was also validated by the actual OPM-MEG evoked experiments of three subjects. Comparative analyses were conducted against the FastICA, Infomax, and AMUSE algorithms. Evaluation metrics included normalized mean square error, normalized delta band power error, RMS error, and signal-to-noise ratio, demonstrating that the proposed method provides optimal suppression of metal artifacts. This advancement holds promise for enhancing data quality and expanding the clinical applications of OPM-MEG.

Removal of Ocular and Muscular Artifacts From Multi-Channel EEG Using Improved Spatial-Frequency Filtering.

Over recent decades, electroencephalogram (EEG) has become an essential tool in the field of clinical analysis and neurological disease research. However, EEG recordings are notably vulnerable to artifacts during acquisition, especially in clinical settings, which can significantly impede the accurate interpretation of neuronal activity. Blind source separation is currently the most popular method for EEG denoising, but most of the sources it separates often contain both artifacts and brain activity, which may lead to substantial information loss if handled improperly. In this paper, we introduce a dual-threshold denoising method combining spatial filtering with frequency-domain filtering to automatically eliminate electrooculogram (EOG) and electromyogram (EMG) artifacts from multi-channel EEG. The proposed method employs a fusion of second-order blind identification (SOBI) and canonical correlation analysis (CCA) to enhance source separation quality, followed by adaptive threshold to localize the artifact sources, and strict fixed threshold to remove strong artifact sources. Stationary wavelet transform (SWT) is utilized to decompose the weak artifact sources, with subsequent adjustment of wavelet coefficients in respective frequency bands tailored to the distinct characteristics of each artifact. The results of synthetic and real datasets show that our proposed method maximally retains the time-domain and frequency-domain information in the EEG during denoising. Compared with existing techniques, the proposed method achieves better denoising performance, which establishes a reliable foundation for subsequent clinical analyses.

Research on Ocular Artifacts Removal from Single-Channel Electroencephalogram Signals in Obstructive Sleep Apnea Patients Based on Support Vector Machine, Improved Variational Mode Decomposition, and Second-Order Blind Identification.

The electroencephalogram (EEG) has recently emerged as a pivotal tool in brain imaging analysis, playing a crucial role in accurately interpreting brain functions and states. To address the problem that the presence of ocular artifacts in the EEG signals of patients with obstructive sleep apnea syndrome (OSAS) severely affects the accuracy of sleep staging recognition, we propose a method that integrates a support vector machine (SVM) with genetic algorithm (GA)-optimized variational mode decomposition (VMD) and second-order blind identification (SOBI) for the removal of ocular artifacts from single-channel EEG signals. The SVM is utilized to identify artifact-contaminated segments within preprocessed single-channel EEG signals. Subsequently, these signals are decomposed into variational modal components across different frequency bands using the GA-optimized VMD algorithm. These components undergo further decomposition via the SOBI algorithm, followed by the computation of their approximate entropy. An approximate entropy threshold is set to identify and remove components laden with ocular artifacts. Finally, the signal is reconstructed using the inverse SOBI and VMD algorithms. To validate the efficacy of our proposed method, we conducted experiments utilizing both simulated data and real OSAS sleep EEG data. The experimental results demonstrate that our algorithm not only effectively mitigates the presence of ocular artifacts but also minimizes EEG signal distortion, thereby enhancing the precision of sleep staging recognition based on the EEG signals of OSAS patients.

A hybrid output-only scheme for precise modal estimation and uncertainty quantification of large-scale structure through vibration-based measurements

In structural health monitoring or operational modal analysis, owing to unknown or unmeasured ambient excitation information, only structural responses are typically available for modal identification. In this context, bias and variance (uncertainty) errors inevitably exist in modal estimates (especially damping estimates), resulting in inaccurate determinations of structural dynamic properties. To this end, a hybrid output-only scheme based on the second-order blind identification, block bootstrap, and enhanced covariance-driven stochastic subspace identification algorithm with first-order perturbation is presented in this paper to perform precise modal estimation and uncertainty quantification. The accuracy and effectiveness of the hybrid scheme in performing modal identification are verified by numerical simulation study of a framework structure. Furthermore, the hybrid method is applied to analyze recorded acceleration responses of a supertall building with 600 m height during Typhoon Kompasu by considering different sensor setups to verify its applicability and robustness in practical applications, and further investigates the dynamic properties of the skyscraper and their variations during the typhoon. The numerical simulation and field measurement studies demonstrate that the hybrid scheme can effectively reduce the bias errors and quantify the variance errors in modal estimates through a single ambient vibration test, thereby improving the confidence and accuracy of modal identification. Notably, the hybrid method is an effective tool to perform high-accuracy modal estimation and uncertainty quantification for large-scale structures under ambient excitations.

Enhancing second-order blind identification for underdetermined operational modal analysis through bandlimited source separation

Blind source separation (BSS) has been demonstrated to be a promising method for operational modal analysis (OMA). However, the conventional BSS-based modal identification methods can only handle the determined or overdetermined identification where the number of sensors must equal or exceed that of active modes, limiting their wider application. Furthermore, the existing methods for underdetermined BSS require sparsity assumption of sources, optimization of model parameters, or structural prior information, undermining the non-parametric and “blind” properties of BSS. This paper proposes a novel method for underdetermined OMA. Firstly, the bandlimited source separation using sinc-dictionaries is performed to efficiently transform the underdetermined problem into several determined or overdetermined ones. An improved second-order blind identification (SOBI) involving Hilbert transform and random time-lagged covariances is then applied for each bandlimited signal to recover modal responses and mode shapes. To attenuate noise, an adaptive singular spectrum analysis (SSA) can be optionally applied to the recovered modal responses for the highly noisy systems. Finally, the natural frequencies and damping ratios are estimated by the single-degree-of-freedom (SDOF) fitting technique. Several numerical examples, an experimental frame case, and a field test of a pedestrian bridge are studied to verify the effectiveness of the proposed strategy.

Tracking gaze position from EEG: Exploring the possibility of an EEG-based virtual eye-tracker.

Ocular artifact has long been viewed as an impediment to the interpretation of electroencephalogram (EEG) signals in basic and applied research. Today, the use of blind source separation (BSS) methods, including independent component analysis (ICA) and second-order blind identification (SOBI), is considered an essential step in improving the quality of neural signals. Recently, we introduced a method consisting of SOBI and a discriminant and similarity (DANS)-based identification method, capable of identifying and extracting eye movement-related components. These recovered components can be localized within ocular structures with a high goodness of fit (>95%). This raised the possibility that such EEG-derived SOBI components may be used to build predictive models for tracking gaze position. As proof of this new concept, we designed an EEG-based virtual eye-tracker (EEG-VET) for tracking eye movement from EEG alone. The EEG-VET is composed of a SOBI algorithm for separating EEG signals into different components, a DANS algorithm for automatically identifying ocular components, and a linear model to transfer ocular components into gaze positions. The prototype of EEG-VET achieved an accuracy of 0.920° and precision of 1.510° of a visual angle in the best participant, whereas an average accuracy of 1.008°±0.357° and a precision of 2.348°±0.580° of a visual angle across all participants (N=18). This work offers a novel approach that readily co-registers eye movement and neural signals from a single-EEG recording, thus increasing the ease of studying neural mechanisms underlying natural cognition in the context of free eye movement.

Blind recovery of sources for multivariate space-time random fields

With advances in modern worlds technology, huge datasets that show dependencies in space as well as in time occur frequently in practice. As an example, several monitoring stations at different geographical locations track hourly concentration measurements of a number of air pollutants for several years. Such a dataset contains thousands of multivariate observations, thus, proper statistical analysis needs to account for dependencies in space and time between and among the different monitored variables. To simplify the consequent multivariate spatio-temporal statistical analysis it might be of interest to detect linear transformations of the original observations that result in straightforward interpretative, spatio-temporally uncorrelated processes that are also highly likely to have a real physical meaning. Blind source separation (BSS) represents a statistical methodology which has the aim to recover so-called latent processes, that exactly meet the former requirements. BSS was already successfully used in sole temporal and sole spatial applications with great success, but, it was not yet introduced for the spatio-temporal case. In this contribution, a reasonable and innovative generalization of BSS for multivariate space-time random fields (stBSS), under second-order stationarity, is proposed, together with two space-time extensions of the well-known algorithms for multiple unknown signals extraction (stAMUSE) and the second-order blind identification (stSOBI) which solve the formulated problem. Furthermore, symmetry and separability properties of the model are elaborated and connections to the space-time linear model of coregionalization and to the classical principal component analysis are drawn. Finally, the usefulness of the new methods is shown in a thorough simulation study and on a real environmental application.

Removing ocular artifacts from magnetoencephalographic data on naturalistic reading of continuous texts.

Naturalistic reading paradigms and stimuli consisting of long continuous texts are essential for characterizing the cortical basis of reading. Due to the highly dynamic nature of the reading process, electrophysiological brain imaging methods with high spatial and temporal resolution, such as magnetoencephalography (MEG), are ideal for tracking them. However, as electrophysiological recordings are sensitive to electromagnetic artifacts, data recorded during naturalistic reading is confounded by ocular artifacts. In this study, we evaluate two different pipelines for removing ocular artifacts from MEG data collected during continuous, naturalistic reading, with the focus on saccades and blinks. Both pipeline alternatives are based on blind source separation methods but differ fundamentally in their approach. The first alternative is a multi-part process, in which saccades are first extracted by applying Second-Order Blind Identification (SOBI) and, subsequently, FastICA is used to extract blinks. The other alternative uses a single powerful method, Adaptive Mixture ICA (AMICA), to remove all artifact types at once. The pipelines were tested, and their effects compared on MEG data recorded from 13 subjects in a naturalistic reading task where the subjects read texts with the length of multiple pages. Both pipelines performed well, extracting the artifacts in a single component per artifact type in most subjects. Signal power was reduced across the whole cortex in all studied frequency bands from 1 to 90 Hz, but especially in the frontal cortex and temporal pole. The results were largely similar for the two pipelines, with the exception that SOBI-FastICA reduced signal in the right frontal cortex in all studied frequency bands more than AMICA. However, there was considerable interindividual variation in the effects of the pipelines. As a holistic conclusion, we choose to recommend AMICA for removing artifacts from MEG data on naturalistic reading but note that the SOBI-FastICA pipeline has also various favorable characteristics.

X-ray fluorescence scanning of soft and wet-sediment cores in terrestrial environments; A robust blind source separation approach

X-ray fluorescence (XRF) scanning of split cores using automated cores scanners is rapid but expensive and semi-quantitative method in sedimentary geochemistry. While searching for a low-cost and quantitative solution in environmental geochemistry, we investigated the application of hand-held XRF device in scanning on two siliciclastic-sediment cores from dam-reservoirs in Slovakia. The core-scan data were compared with XRF data from dry, powdered sample aliquots, grain size, water content, and magnetic susceptibility data, which were used as independent variables. The aim of the manuscript is to investigate the applicability of time-series statistical methods to reveal the physical and chemical signals from hand-held XRF scanning of wet-sediment cores. The geochemical data were processed using the centred log-ratio (clr) technique, and then by the Second-Order Blind Identification (SOBI) method, which is a subtype of robust Blind Source Separation (BSS) techniques commonly used to enhance the signal-to-noise ratio in time series analysis of data in medicine or acoustics. So far, BSS has been rarely applied in geochemistry. Raw element concentrations (ppm/wt%) of core scans and sediment powders are essentially incomparable, but their clr coefficients show statistically significant correlation for Al, P, K, Ca, Mn, Fe, Cu, As, Sr, Zr, and Pb, which reduces the difference between the two measurement methods and justifies previous attempts to calibrate the core-scanning data using log-ratio techniques. Scores of three of eight latent components (IC2, IC4, and IC7) found by SOBI in the core scan- and powder data show significant statistical correlation demonstrating the applicability of the core-scan technique. Their geochemical interpretation is based on the IC loadings by element clr, and the stratigraphic correlation with the sediment grain size and water content. These ICs explain geochemical variability due to grain-size contrast between sandy and silty layers, input of biogenic productivity-sensitive elements (Ca and P), eutrophication, and accumulation of anthropogenic elements. The robust BSS approach represents a promising method of processing of large datasets in environmental geochemistry.

A robust methodology for output-only modal identification of civil engineering structures

Output-only modal identification has been attracting increasing research efforts in civil engineering community. The aim of this study was to develop a robust output-only modal identification methodology to improve the performance of the existing methods. The proposed method combines the second-order blind identification (SOBI) with the covariance driven stochastic subspace identification (SSI-COV) by introducing a new system model. Instead of employing single degree-of-freedom fitting methods to obtain local modal information of each mode like it usually does in SOBI, modal responses identified by SOBI are regarded as the output of a system to form the block Toeplitz matrix of the SSI-COV. More accurate global modal parameters are expected to be obtained since a parametric system model is included and the contributions of different modes are considered. It is also easy to select physical modes based on the time and frequency characteristics of source signals and thus can avoid the ambiguous work related to the stabilization diagrams. The proposed method is validated by a simulated 8-DOF system with emphasize on identification accuracy and parameter sensitivity. The results confirm the insensitivity of input parameters and demonstrate the high accuracy. It is then applied to estimate the modal parameters of a reinforced concrete shear core building and a concrete arch-gravity dam. It is demonstrated that a total of twelve building modes and six dam modes are successfully identified respectively, and the results in both applications are accord well with the existing modal identification methods, especially SSI-COV. The promising results show that the physical modes can be well separated, and the associated modal parameters can be reliably identified with high accuracy, even if closely-spaced modes and spurious harmonics occur. As expected, the modes discrimination step that associated with the proposed method tends to be more distinct and intuitive than the stabilization diagrams.

Remove Artifacts from a Single-Channel EEG Based on VMD and SOBI.

With the development of portable EEG acquisition systems, the collected EEG has gradually changed from being multi-channel to few-channel or single-channel, thus the removal of single-channel EEG signal artifacts is extremely significant. For the artifact removal of single-channel EEG signals, the current mainstream method is generally a combination of the decomposition method and the blind source separation (BSS) method. Between them, a combination of empirical mode decomposition (EMD) and its derivative methods and ICA has been used in single-channel EEG artifact removal. However, EMD is prone to modal mixing and it has no relevant theoretical basis, thus it is not as good as variational modal decomposition (VMD) in terms of the decomposition effect. In the ICA algorithm, the implementation method based on high-order statistics is widely used, but it is not as effective as the implementation method based on second order statistics in processing EMG artifacts. Therefore, aiming at the main artifacts in single-channel EEG signals, including EOG and EMG artifacts, this paper proposed a method of artifact removal combining variational mode decomposition (VMD) and second order blind identification (SOBI). Semi-simulation experiments show that, compared with the existing EEMD-SOBI method, this method has a better removal effect on EOG and EMG artifacts, and can preserve useful information to the greatest extent.

EEG artifact removal using sub-space decomposition, nonlinear dynamics, stationary wavelet transform and machine learning algorithms.

Blind source separation (BSS) methods have received a great deal of attention in electroencephalogram (EEG) artifact elimination as they are routine and standard signal processing tools to remove artifacts and reserve desired neural information. On the other hand, a classifier should follow BSS methods to automatically identify artifactual sources and remove them in the following steps. In addition, removing all detected artifactual components leads to loss of information since some desired information related to neural activity leaks to these sources. So, an approach should be employed to detect and suppress the artifacts and reserve neural activity. This study introduces a novel method based on EEG and Poincare planes in the phase space to detect artifactual components estimated by second-order blind identification (SOBI). Artifacts are detected using a mixture of well-known conventional classifiers and were removed employing stationary wavelet transform (SWT) to reserve neural information. The proposed method is a combination of signal processing techniques and machine learning algorithms, including multi-layer perceptron (MLP), K-nearest neighbor (KNN), naïve Bayes, and support vector machine (SVM) which have significant results while applying our proposed method to different scenarios. Simulated, semi-simulated, and real EEG signals are employed to evaluate the proposed method, and several evaluation criteria are calculated. We achieved acceptable results, for example, 98% average accuracy and 97% average sensitivity in artifactual EEG component detection or about 2% as mean square error in EEG reconstruction after artifact removal. Results showed that the proposed method is effective and can be used in future studies as we have considered different real-world scenarios to evaluate it.

Performance Evaluation of Blind Modal Identification in Large-Scale Civil Infrastructure

The monitoring and maintenance of existing civil infrastructure has recently received worldwide attention. Several structural health monitoring methods have been developed, including time-, frequency-, and time–frequency domain methods of modal identification and damage detection to estimate the structural and modal parameters of large-scale structures. However, there are several implementation challenges of these modal identification methods, depending on the size of the structures, measurement noise, number of available sensors, and their operational loads. In this paper, two modal identification methods, Second-Order Blind Identification (SOBI) and Time-Varying Filtering Empirical Mode Decomposition (TVF-EMD), are evaluated and compared for large-scale structures including a footbridge and a wind turbine blade with a wide range of dynamic characteristics. The results show that TVF-EMD results in better accuracy in modal identification compared to SOBI for both structures. However, when the number of sensors is equal to or more than the number of target modes of the structure, SOBI results in better computational efficiencies compared to TVF-EMD.

A resting-state network for novelty: Similar involvement of a global network under rest and task conditions

Neuroimaging research provides converging evidence in support of functional networks active under rest conditions. While these networks are typically locally-distributed, a globally-distributed resting-state network (gRSN) was recently identified. The gRSN component is characterized by a scalp topography similar to that of the widely-studied P3 component of the event related potential, thought to represent the brain's response to novelty. In this study, we investigate similarities between the neural generators underlying these two networks to test the hypothesis that the gRSN is a resting-state network for novelty. By using the second-order blind identification (SOBI) algorithm, which works with temporal information, we show that (1) a resting-state component resembling the topography of the P3 can be recovered in all participants; (2) this gRSN component can be modeled with a set of ECDs with high goodness of fit; (3) ECD locations of the gRSN correspond to a network of globally-distributed brain structures overlapping heavily with the networking underlying the P3; and, (4) structures underlying these two networks are similarly involved during task and resting-state conditions. We interpret this as evidence in support of a resting-state network for detection and response to novelty.

A Blind Filtering Framework for Noisy Neonatal Chest Sounds

Chest sound— as the first and most commonly available vital signal for newborns— contains affluent information about their cardiac and respiratory health. However, neonatal lung sound auscultation is currently challenging and often unreliable due to the noise and interference, particularly for preterm infants. The noise often overlaps with the heart and lung contents in both time and frequency. Moreover, the frequency band of the useful components varies from one case to another, making it difficult to separate by fixed band-pass filtering. In this study, a single-channel Blind Source Separation (SCBSS) framework is proposed to separate newborns’ lung and heart sounds from noisy chest sounds recorded by a digital stethoscope. This method first decomposes the signal into a multi-resolution representation using a time-frequency transform, and then applies source separation algorithms, to find proper ad hoc frequency filters. In the simulation scenario, two different time-frequency transforms are considered; Stationary Wavelet Transform (SWT) with dyadic bases, and Continuous Wavelet Transform (CWT) with redundant bases. The transforms are followed by three different source separation methods, namely Principal Component Analysis (PCA), Periodic Component Analysis (πCA), and Second Order Blind Identification (SOBI). The yielded combinations are applied to the chest sounds recorded from ninety-one preterm and full-term newborns. The results show that compared to raw signals, fixed band-pass filtering and seven other separation methods, the heart and lung sounds extracted by the proposed methods have higher quality index and also result in more reliable heart and respiratory rate estimation.