Evaluation of data-driven MEG analysis approaches for the extraction of fMRI resting state networks

The electrophysiological basis of resting state networks (RSN) is still under debate. In particular, no principled mechanism has been determined that is capable of explaining all RSN equally well. While magnetoencephalography (MEG) and electroencephalography (EEG) are the methods of choice to determine the electrophysiological basis of RSN, no standard analysis pipeline of RSN yet exists. In this paper, we compare the two main existing data-driven analysis strategies for extracting resting state networks from MEG data. The first approach extracts RSN through an independent component analysis (ICA) of the Hilbert envelope in different frequency bands. The second approach uses phase –amplitude coupling to determine the RSN. To evaluate the performance of these approaches, we compare the MEG-RSN to the functional magnetic resonance imaging (fMRI)-RSN from the same subjects.Overall, it was possible to extract the canonical fMRI RSN with MEG. The approach based on phase-amplitude coupling yielded the best correspondence to the fMRI-RSN. The Hilbert envelope-ICA produced different dominant frequency-bands underlying RSN for different ICA runs, suggesting the absence of a single dominant frequency underlying the RSN. Our results also suggest that individual RSN are not characterized by one single dominant frequency. Instead, the resting state networks seem to be based on a combination of the delta/theta phase and gamma amplitude.

Detection of broken rotor bars of induction motors based on the combination of Hilbert envelope analysis and Shannon entropy

Broken rotor bar (BRB) is one of the most common fault types of induction motors. One of the common methods to detect the broken rotor bars is the observation of the characteristic sideband frequencies in the stator current. If the motor is lightly loaded, the sideband harmonics are attached to the fundamental frequency of the main supply and the amplitudes of these harmonics are quite low. Therefore, it is difficult to detect the broken rotor bars under light loading conditions by using conventional motor current signature analysis (MCSA) methods. Moreover, in some cases, the sideband harmonics of fundamental frequency may exist although there is no rotor fault in induction motors due to load oscillations. Therefore, there is a risk for false broken rotor bars alarm with the existence of lower amplitude of harmonics. This paper provides an alternative approach for the detection of broken rotor bars by applying Hilbert envelope analysis along with Shannon entropy to stator current signals. The proposed method includes two-stage evaluation system to eliminate false BRB alarms such as detecting sidebands from envelope spectrum and calculating entropy rates from envelope signals. The results are verified experimentally under 25 %, 50 %, 75 % and 100 % loading conditions.

Rolling Bearing Fault Diagnosis based on Residual Neural Network

<div class="Section0"><div>Because rolling bearings have been working in an environment with complex and variable working conditions and large noise interference for a long time, the bearing fault diagnosis method has a poor diagnostic effect under variable working conditions. To solve this problem, we propose a residual neural network based on the diagnosis method of rolling bearing fault. The proposed method takes rolling bearing time-domain signal data as input. Because bearing signals have strong time-varying properties, we construct a multi-scale residual block that can not only learn features at different levels, but also expand the width and depth of the residual neural network. We use the advantages of the dilated convolution to expand the receptive field, replace part of the ordinary convolution in the multi-scale residual block with the dilated convolution, and design a multi-scale hollow residual block. The advantage is that the method is made by expanding the receptive field. It has a strong feature learning ability and can learn better features under limited data. Finally, we add a Dropout layer to discard a certain proportion of neurons after the fully connected layer, which can effectively avoid the negative impact of overfitting, and use Case Western Reserve University bearing dataset, the simulation experiment, and the SVM + EMD + Hilbert envelope spectrum, BPNN + EMD + Hilbert envelope spectrum and Resnet three ways of comparative analysis, the results show that the method under the variable condition of the fault diagnosis of rolling bearing has higher diagnosis accuracy, stronger noise resistance, and generalization ability.</div><p> </p></div><p> </p>

Hilbert-Envelope Features for Cardiac Disease Classification from Noisy Phonocardiograms

Cardiovascular disease (CVD) is considered a significant public health concern around the world. Automated early diagnostic tools for CVDs can provide substantial benefits, especially in low-resource countries. In this study, we propose a time-domain Hilbert envelope feature (HEF) extraction scheme that can effectively distinguish among different cardiac anomalies from heart sounds even in highly noisy recordings. The method is motivated by how a cardiologist listens to the heart murmur configuration, e.g., the intensity of the heart sound envelope over a cardiac cycle. The proposed feature is invariant to the heart rate, the position of the first and second heart sounds, and robust in extracting the murmur configuration pattern in the presence of respiratory noise. Experimental evaluations are performed compared to two different state-of-the-art methods in the presence of respiratory noise with signal-to-noise ratio (SNR) values ranging from 0-15dB. The proposed HEF, fused with standard acoustic and Resnet features, yields an average accuracy, sensitivity, specificity, and F1-score of, 94.78%(±2.63), 87.48%(±6.07), 96.87%(±1.51) and 87.47%(±5.94), respectively, while using a random forest (RF) classifier. Compared to the best-performing baseline model, this feature-fusion scheme provides a significant performance improvement (p < 0.05), notably achieving an absolute improvement of 6.16% in averaged sensitivity. In the case of 0dB SNR, the proposed feature alone provides a 9.2% absolute improvement in sensitivity compared to the top baseline system demonstrating the robustness of the HEF. The developed methods can significantly impact computer-aided auscultation (CAA) systems when deployed in noisy conditions, especially in low-resource settings.

An Enhanced VMD with the Guidance of Envelope Negentropy Spectrum for Bearing Fault Diagnosis

Currently, study on the relevant methods of variational mode decomposition (VMD) is mainly focused on the selection of the number of decomposed modes and the bandwidth parameter using various optimization algorithms. Most of these methods utilize the genetic-like algorithms to quantitatively analyze these parameters, which increase the additional initial parameters and inevitably the computational burden due to ignoring the inherent characteristics of the VMD. From the perspective to locate the initial center frequency (ICF) during the VMD decomposition process, we propose an enhanced VMD with the guidance of envelope negentropy spectrum for bearing fault diagnosis, thus effectively avoiding the drawbacks of the current VMD-based algorithms. First, the ICF is coarsely located by envelope negentropy spectrum (ENS) and the fault-related modes are fast extracted by incorporating the ICF into the VMD. Then, the fault-related modes are adaptively optimized by adjusting the bandwidth parameters. Lastly, in order to identify fault-related features, the Hilbert envelope demodulation technique is used to analyze the optimal mode obtained by the proposed method. Analysis results of simulated and experimental data indicate that the proposed method is effective to extract the weak faulty characteristics of bearings and has advantage over some advanced methods. Moreover, a discussion on the extension of the proposed method is put forward to identify multicomponents for broadening its applied scope.

Improving the Accuracy of Fault Frequency by Means of Local Mean Decomposition and Ratio Correction Method for Rolling Bearing Failure

The fault frequencies are as they are and cannot be improved. One can only improve its estimation quality. This paper proposes a fault diagnosis method by combining local mean decomposition (LMD) and the ratio correction method to process the short-time signals. Firstly, the vibration signal of rolling bearing is decomposed into a series of product functions (PFs) by LMD. The PF, which contains the richest fault information, is selected to perform envelope spectrum analysis by the Hilbert transform (HT). Secondly, the Hilbert envelope spectrum of the selected PF is corrected with the ratio correction method. Finally, higher precision fault frequencies are extracted from the corrected Hilbert envelope spectrum, and then the fault location is accurately determined. The proposed method of this paper can be used in online real-time monitoring technology of rolling bearing failure.