Automated Dynamic Electrocardiogram Noise Reduction Using Multilayer LSTM Network

Time-frequency masking is a method for noise reduction that is based on the time-frequency representation of a speech in noise signal. Depending on the estimated signal-to-noise ratio (SNR), each time-frequency unit is either attenuated or not. A special type of a time-frequency mask is the ideal binary mask (IBM), which has access to the real SNR (ideal). The IBM either retains or removes each time-frequency unit (binary mask). The IBM provides large improvements in speech intelligibility and is a valuable tool for investigating how different factors influence intelligibility. This study extends the standard outcome measure (speech intelligibility) with additional perceptual measures relevant for noise reduction: listening effort, noise annoyance, speech naturalness, and overall preference. Four types of time-frequency masking were evaluated: the original IBM, a tempered version of the IBM (called ITM) which applies limited and non-binary attenuation, and non-ideal masking (also tempered) with two different types of noise-estimation algorithms. The results from ideal masking imply that there is a trade-off between intelligibility and sound quality, which depends on the attenuation strength. Additionally, the results for non-ideal masking suggest that subjective measures can show effects of noise reduction even if noise reduction does not lead to differences in intelligibility.

An application of the wavelet transform to electrocardiography is described in the paper. To detect the weak signals buried in noises is a fundamental and important problem. The proposed technique is based on the decomposition of the ECG signal into a set of basic functions covering the time- frequency domain. It has been found that digital filters are not suitable for processing weak signals in noises, while discrete wavelet transform (DWT) is used to analyze weak digital signal and extract small-features. In this paper, a new procedure is proposed, called sub-band filter. The performance for our method is investigated by detecting the ECG signal in white noise. The results show that the DTW is a quite effective method for the extraction features of weak signal and improving the ratio of signal to noise.

The link between oral diseases and mortality remains under-explored. We aimed to evaluate the associations between tooth count, untreated caries and risk of all-cause and cause-specific mortality. Data on 24 029 adults from the National Health and Nutrition Examination Survey 1988-94/1999-2010, with mortality linkage to the National Death Index to 31 December 2015, were analysed. Baseline total number of permanent teeth and any untreated caries were assessed by trained dental professionals. During up to 27 years of follow-up, 5270 deaths occurred. Fewer permanent teeth were associated with higher all-cause mortality, including heart disease and cancer mortality (all P <0.05 for trend) but not cerebrovascular disease mortality. For every 10 teeth missing, the multivariable-adjusted hazard ratios (HRs) were 1.13 (95% CI: 1.08 to 1.18) for all-cause, 1.16 (95% CI: 1.05, 1.29) for heart disease and 1.19 (95% CI: 1.09, 1.29) for cancer mortality. Untreated caries was associated with increased all-cause (HR: 1.26, 95% CI: 1.15, 1.39) and heart disease mortality (HR: 1.48, 95% CI: 1.17, 1.88) but not cerebrovascular disease/cancer mortality, after adjusting for tooth count, periodontitis and sociodemographic/lifestyle factors. Compared with those without untreated caries and with 25-28 teeth, individuals with untreated caries and 1-16 teeth had a 53% increased risk of all-cause mortality (HR: 1.53, 95% CI: 1.27, 1.85) and 96 % increased risk of heart disease mortality (HR: 1.96, 95% CI: 1.28, 3.01). In nationally representative cohorts, fewer permanent teeth and untreated caries were associated with all-cause and heart disease mortality. Fewer teeth were also associated with higher cancer mortality.

Currently, robots on the market have varying operational requirements due to their different functions. The care robots are primarily used to serve people under care. Therefore, reducing the amount of noise they make while moving is important. This study measured their noise in an anechoic room. This study found that the most noise is produced when they move forward and backward at full speed. The robot’s vibration signal was measured and compared with the noise source to propose the most effective way to reduce the noise. Due to the robot's small internal space, methods for noise reduction are limited. The care robot’s development must adopt quick customizable 3D printing technologies, produce sound absorption/insulation materials independently, determine the best sound absorption/insulation material and structural configuration, use impedance tubes to measure the sound absorption coefficient and sound insulation performance, and select the appropriate materials and implement them into the robot for actual tests. These results indicate that the noise value of the main contributing frequency can be reduced by 15dB at most. Furthermore, this study also used an acoustic simulation analysis to calculate and simulate the vibration and noise of the care robot. The vibration and noise signals during the robot's operation were measured and compared to provide a reference for manufacturers in design. Finally, experimental materials were added to the simulation and compared with the noise reduction results. The results show similar trends. Therefore, the model developed in this study may provide a fast, low-cost, and feasible method for the noise reduction of care robots.

Problem/ConditionHeart disease is the leading cause of death in the United States. In 2015, heart disease accounted for approximately 630,000 deaths, representing one in four deaths in the United States. Although heart disease death rates decreased 68% for the total population from 1968 to 2015, marked disparities in decreases exist by race and state.Period Covered1968–2015.Description of SystemThe National Vital Statistics System (NVSS) data on deaths in the United States were abstracted for heart disease using diagnosis codes from the eighth, ninth, and tenth revisions of the International Classification of Diseases (ICD-8, ICD-9, and ICD-10) for 1968–2015. Population estimates were obtained from NVSS files. National and state-specific heart disease death rates for the total population and by race for adults aged ≥35 years were calculated for 1968–2015. National and state-specific black-white heart disease mortality ratios also were calculated. Death rates were age standardized to the 2000 U.S. standard population. Joinpoint regression was used to perform time trend analyses.ResultsFrom 1968 to 2015, heart disease death rates decreased for the total U.S. population among adults aged ≥35 years, from 1,034.5 to 327.2 per 100,000 population, respectively, with variations in the magnitude of decreases by race and state. Rates decreased for the total population an average of 2.4% per year, with greater average decreases among whites (2.4% per year) than blacks (2.2% per year).At the national level, heart disease death rates for blacks and whites were similar at the start of the study period (1968) but began to diverge in the late 1970s, when rates for blacks plateaued while rates for whites continued to decrease. Heart disease death rates among blacks remained higher than among whites for the remainder of the study period. Nationwide, the black-white ratio of heart disease death rates increased from 1.04 in 1968 to 1.21 in 2015, with large increases occurring during the 1970s and 1980s followed by small but steady increases until approximately 2005. Since 2005, modest decreases have occurred in the black-white ratio of heart disease death rates at the national level. The majority of states had increases in black-white mortality ratios from 1968 to 2015. The number of states with black-white mortality ratios >1 increased from 16 (40%) to 27 (67.5%). InterpretationAlthough heart disease death rates decreased both for blacks and whites from 1968 to 2015, substantial differences in decreases were found by race and state. At the national level and in most states, blacks experienced smaller decreases in heart disease death rates than whites for the majority of the period. Overall, the black-white disparity in heart disease death rates increased from 1968 to 2005, with a modest decrease from 2005 to 2015.Public Health ActionSince 1968, substantial increases have occurred in black-white disparities of heart disease death rates in the United States at the national level and in many states. These increases appear to be due to faster decreases in heart disease death rates for whites than blacks, particularly from the late 1970s until the mid-2000s. Despite modest decreases in black-white disparities at the national level since 2005, in 2015, heart disease death rates were 21% higher among blacks than among whites. This study demonstrates the use of NVSS data to conduct surveillance of heart disease death rates by race and of black-white disparities in heart disease death rates. Continued surveillance of temporal trends in heart disease death rates by race can provide valuable information to policy makers and public health practitioners working to reduce heart disease death rates both for blacks and whites and disparities between blacks and whites.

Noise reduction for speech applications is often formulated as a digital filtering problem, where the clean speech estimate is obtained by passing the noisy speech through a linear filter/transform. With such a formulation, the core issue of noise reduction becomes how to design an optimal filter (based on the statistics of the speech and noise signals) that can significantly suppress noise without introducing perceptually noticeable speech distortion. The optimal filters can be designed either in the time or in a transform domain. The advantage of working in a transform space is that, if the transform is selected properly, the speech and noise signals may be better separated in that space, thereby enabling better filter estimation and noise reduction performance. Although many different transforms exist, most efforts in the field of noise reduction have been focused only on the Fourier and Karhunen-Loeve transforms. Even with these two, no formal study has been carried out to investigate which transform can outperform the other. In this paper, we reformulate the noise reduction problem into a more generalized transform domain. We will show some of the advantages of working in this generalized domain, such as 1) different transforms can be used to replace each other without any requirement to change the algorithm (optimal filter) formulation, and 2) it is easier to fairly compare different transforms for their noise reduction performance. We will also address how to design different optimal and suboptimal filters in such a generalized transform domain.

Acoustic environment has a low signal-to-noise ratio (SNR); hence, array signal processing is widely used for noise reduction and signal enhancement. The actual ambient noise includes uncorrelated noise and correlated noise. The received noises of the two arbitrary array elements are correlated. Consequently, the performance of array signal processing method decreases obviously. Aiming at this problem, the real-part elimination of covariance matrix method is proposed. Firstly, from a physical point of view, the noise signals can be generated by using a number of uncorrelated noise sources: the more the noise sources, the less the error between the noise from the model and the actual noise will be. Theoretically, the noise field is decomposed into the symmetrical noise field and the asymmetrical noise field. A number of noise sources generate the symmetrical noise fields; the directions of these noise sources are symmetric, and the powers of two arbitrary symmetric sources are the same. Secondly, the symmetry of the ambient noise is analyzed, as a result, the symmetrical noise can only affect the real part of the covariance matrix. Thirdly, the real part of covariance matrix is eliminated in order to reduce the noise, and then the delay-and-sum beamforming is achieved by using only the imaginary part. The advantages are that the output signal-to-noise ratio is increased and the noise output power is reduced obviously; the disadvantage is that it produces a false target. The azimuth of the actual target differs from that of the false target by 180, and the false target cannot be distinguished. Finally, to eliminate the false target, the real part of the signal covariance matrix is reconstructed by establishing a constrained optimization problem, which is solved by using the particle swarm algorithm. Then, the reconstructed covariance matrix composed of the imaginary part and the reconstruction of real part is applied to delay-and-sum beamforming, as a result, the false target is eliminated. The simulation results show that the real-part elimination of covariance matrix method reduces the symmetrical ambient noise, the noise output power is reduced, the output signal-to-noise ratio is increased, and this method improves the performance of array signal processing. The experimental results show that the output SNRs of two targets with using the imaginary part of covariance matrix are increased by 3.57 dB and 3.149 dB, respectively, and the output SNRs of two targets with using the reconstructed covariance matrix are increased by 7.027 dB and 6.985 dB, respectively. The real-part elimination of covariance matrix method is easy to implemente, and has a definite value for engineering application.

The presence of baseline wander in ECG signal severely affects the quality of ECG signal. Baseline wander is generally eliminated at the initial stage of preprocessing of ECG signal. In this paper a method using regression estimation is applied to ECG signal to remove the baseline wander. This method uses two stage median filter and regression estimation for smoothing and correction of baseline. The cross correlation is performed between the baseline wander containing ECG signal with corrected ECG signal. Then cross-correlation coefficient is calculated. The method evolves with the lowess and loess of local regression estimation, rlowess and rloess of robust local regression estimation. We found highest coefficient of 0.9914 for the baseline wander amplitude of 0.5mV using local regression method (lowess) and lowest coefficient of 0.9688 for the amplitude of 0.8mV using robust local regression method (rlowess). Further this method is validated on MIT-BIH, Noise Stress Test Database (NSTD).

Recently, with the rapid growth of wearable healthcare devices, there has been an increasing demand for transparent and flexible electrode technologies that can adhere stably to curved and irregular surfaces while maintaining signal stability during long-term measurements. In this study, we propose a transparent and flexible CNT–PDMS based micro-column electrode that incorporates microscale protrusions to enable multi-point contact, thereby reducing contact impedance and enhancing signal stability. The proposed electrode features a continuous monolithic structure integrating an upper micro-column layer with a lower grid framework, which expands the effective contact area and uniformly distributes contact pressure on curved surfaces. CNT content optimization experiments (5–25 wt%) confirmed that a stable conductive pathway was established at 20 wt%, which was selected as the final composition. ECG signal quality was comparatively evaluated under identical conditions using commercial Ag/AgCl electrodes (gel removed), sheet-type electrodes, and micro-column electrodes with diameters of 480 μm, 690 μm, and 1,020 μm. The 690 μm micro-column electrode exhibited the best performance, with RMS 0.75% higher and SNR 4.7 dB greater than the commercial Ag/AgCl electrode. These results demonstrate that stable ECG signal acquisition can be achieved without conductive gel in both initial and long-term measurements, highlighting the electrode’s high applicability in curved-surface environments, wearable devices, and in-vehicle biosignal monitoring applications.

Automatic heart disease detection from human heartbeats is a challenging and intellectual assignment in signal processing because periodically monitoring of the heart beat arrhythmia for patient is an essential task to reduce the death rate due to cardiovascular disease (CVD). In this paper, the focus of research is to design hybrid Convolutional Neural Network (CNN) architecture by making use of Grasshopper Optimization Algorithm (GOA) to classify different types of heart diseases from the ECG signal or human heartbeats. Convolutional Neural Network (CNN) as an artificial intelligence approach is widely used in computer vision-based medical data analysis. However, the traditional CNN cannot be used for classification of heart diseases from the ECG signal because lots of noise or irrelevant data is mixed with signal. So this study utilizes the pre-processing and selection of feature for proper heart diseases classification, where Discrete Wavelet Transform (DWT) is used for the noise reduction as well as segmentation of ECG signal and Grasshopper Optimization Algorithm (GOA) is used for selection of R-peaks features from the extracted feature sets in terms of R-peaks and R-R intervals that help to attain better classification accuracy. For training as well as testing of projected Heartbeats Classification Model (HCM), the Standard MIT-BIH arrhythmia database is utilized with hybrid Convolutional Neural Network (CNN) architecture. The assortment of proper R-peaks and R-R intervals is a major factor and because of the deficiency of apposite pre-processing phases like noise removal, signal decomposition, smoothing and filtering, the uniqueness of extracted feature is less. The experimental outcomes show that the planned HCM is effective for detecting irregular human heartbeats via R-peaks and R-R intervals. When the proposed Heartbeats Classification Model (HCM) was verified on the database, model achieved higher efficiency than other state-of-the-art techniques for 16 heartbeat disease categories and the average classification accuracy is 99.58% with fast and robust responses where the correctly classified heartbeats are 86,005 and misclassified beats is only 108 with 0.42% error rate.

In the present study, a nonlinear filtering technique using median filter is proposed for noise reduction and delineation of standard ECG signal. The advantage of using nonlinear median filter lies in the fact that no mathematical transformation is used. Hence, time and space complexity are also reduced. As the mathematical transformation introduces distortion in the acquired ECG signal, nonlinear median filter is a great choice to denoise the ECG signal. A single-median filtering technique is capable of removing all possible artifacts present in recorded ECG signals. QRS complex is identified from the filtered signal using the same median filtering technique. Result shows the effectiveness of the proposed method of artifact removal from recorded ECG signals. This method is computationally very simple and can be applied in real-time operations without any appreciable time delay due to signal processing stage.KeywordsBaseline wandering artifactECGMedian filterPower line interference artifactQRS detection

Objective:Despite the long-term decline, heart disease has remained the leading cause of death in the United States (US) over the past eight decades, accounting for 23% of all deaths in 2017. Although psychological distress has been associated with cardiovascular disease mortality, the relationship between different psychological distress levels and heart disease mortality in the US has not been analyzed in detail. Using a national longitudinal dataset, we examined the association between levels of psychological distress and US heart disease mortality.Methods:We analyzed the Kessler 6-item psychological distress scale as a risk factor for heart disease mortality using the pooled 1997-2014 data from the National Health Interview Survey (NHIS) linked to National Death Index (NDI) (N=513,081). Cox proportional hazards regression was used to model survival time as a function of psychological distress and sociodemographic and behavioral covariates.Results:In Cox models with 18 years of mortality follow-up, the heart disease mortality risk was 121% higher (hazard ratio [HR]=2.21; 95% CI=1.99,2.45) in adults with serious psychological distress (SPD) (p<0.001), controlling for age, and 96% higher (HR=1.96; 95% CI=1.77,2.18) in adults with SPD (p<0.001), controlling for age, gender, race/ethnicity, immigrant status, education, marital status, poverty status, housing tenure, and geographic region when compared with adults without psychological distress. The relative risk of heart disease mortality associated with SPD decreased but remained significant (HR=1.14, 95% CI=1.02,1.28) after controlling for additional covariates of smoking, alcohol consumption, self-assessed health, activity limitation, and body mass index. There was a dose-response relationship, with relative risks of heart disease mortality increasing consistently at higher levels of psychological distress. Moreover, the association varied significantly by gender and race/ethnicity. The relative risk of heart disease mortality for those who experienced SPD was 2.42 for non-Hispanic Whites and 1.76 for non-Hispanic Blacks, compared with their counterparts who did not experience psychological distress.Conclusions and Global Health Implications:US adults with serious psychological distress had statistically significantly higher heart disease mortality risks than those without psychological distress. These findings underscore the significance of addressing psychological well-being in the population as a strategy for reducing heart disease mortality.

Coronary artery disease (CAD) is the number one killer of adults in the United States, claiming one-half million deaths annually. Early detection and prevention strategies clearly remain a top priority for health care providers in order to reduce the high mortality rate of heart disease. As an unequivocal reflection of arteriosclerosis, coronary arterial calcium (CAC) may provide a means to qualitatively assess the overall disease severity and likewise serve as a means to assess risk for CHD. It is known that patients with heavy calcium burdens have more advanced CAD, a concomitantly a higher likelihood of coronary stenoses, and a concomitant higher risk for acute cardiac events. Computed tomography has been shown to be an accurate, non-invasive method to quantify coronary calcification burden in patients. Evidence shows that calcium measurements by CT correlate well with histological plaque analyses, and that CAC measurements accurately reflect disease severity and can be useful to assess individual risk for CHD. The purpose of this article is to summarize the currently available evidence that has attempted to validate CAC screening as a screening exam and risk predictor for coronary heart disease.

In this chapter, fundamentals of adaptive filter are explained. Application of adaptive filter over EEG and ECG signals has been demonstrated and explained clearly in a simple way. All the implementation details of LMS and NLMS algorithms for adaptive filter are also provided along with MatLab codes. This chapter provides a complete study of noise cancellation of EEG and ECG signals. These studies include dataset description, details about noise signal, and modeling of problem as adaptive noise canceller model. Proper visualization of signal before and after the filtering process is given with values of different fidelity parameters used to check the quality of filtered signals. MatLab code for the complete study is also provided to give the real experience of entire process. Famous EEG and ECG datasets, which are publicly accessible, are used. With the input of this chapter, any reader can easily implement and use adaptive filters for filtering of any biomedical signals.

The existence of noise will make the results of time-frequency analysis of blasting seismic wave lack of authenticity. In order to obtain the real blasting vibration characteristics, a method based on the complementary ensemble empirical mode decomposition with adaptive noise and establishing objective function is proposed. The process of noise reduction is divided into three steps. Firstly, the noise signal is decomposed by the complete ensemble empirical mode decomposition with adaptive noise, and the noise reduction algorithm is established based on the intrinsic mode function obtained from the decomposition; secondly, the objective function considering the smoothness of the noise reduction algorithm and its correlation with the measured signal is established; finally, the algorithm corresponding to the optimal solution of the objective function is found, which is the noise reduction model of blasting seismic wave signal. The model is applied to the de-noising of blasting seismic signal, and the de-noising ability of the model is analyzed by the noise reduction error ratio. The analysis results show that the model can reduce the noise of the seismic monitoring signal with noise on the premise of fully preserving the real components of the blasting seismic signal.