Enhancement of lung sounds based on empirical mode decomposition and Fourier transform algorithm

Mondal, Ashok; Banerjee, P.; Somkuwar, Ajay

doi:10.1016/j.cmpb.2016.10.025

Cited by 25 publications

(11 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is also worth mentioning that LS signals are characterized by nonperiodicity and low frequencies in the range of 50 − 2500 [29]. Heart sound (HS) are also within the frequency range of 20 − 600 , with most of the frequency components falling in the range 20 − 150 [32]. The two signals thus interfere with each other, necessitating the need for separating them before LS could be analysed.…”

Section: Discussionmentioning

confidence: 99%

On the effectiveness of signal decomposition, feature extraction and selection on lung sound classification

Elsetrønning,

Rasheed,

Bekker

et al. 2020

Preprint

View full text Add to dashboard Cite

Lung sounds refer to the sound generated by air moving through the respiratory system. These sounds, as most biomedical signals, are non-linear and non-stationary. A vital part of using the lung sound for disease detection is discrimination between normal lung sound and abnormal lung sound. In this paper, several approaches for classifying between no-crackle and crackle lung sounds are explored. Decomposition methods such as Empirical Mode Decomposition, Ensemble Empirical Mode Decomposition, and Discrete Wavelet Transform are used along with several feature extraction techniques like Principal Component Analysis and Autoencoder, to explore how various classifiers perform for the given task. An open-source dataset downloaded from Kaggle, containing chest auscultation of varying quality is used to determine the results of using the different decomposition and feature extraction combinations. It is found that when higher-order statistical and spectral features along with the Mel-frequency cepstral coefficients are fed to the classier we get the best performance with the -NN classifier giving the best accuracy. Furthermore, it is also demonstrated that using a combination of feature selection methods one can significantly reduce the number of input features without adversely affecting the accuracy of the classifiers.

show abstract

Section: Discussionmentioning

confidence: 99%

On the effectiveness of signal decomposition, feature extraction and selection on lung sound classification

Elsetrønning,

Rasheed,

Bekker

et al. 2020

Preprint

View full text Add to dashboard Cite

show abstract

“…7,[14][15][16] Heart sounds are quasi-periodic signals originated by the blood flow c irculating t hrough t he h eart combined with the movement of its own structure. 4,7,17 Its main components are the first heart sound (S 1 ) generated during ventricular systole (closure of the atrioventricular valves: mitral/bicuspid and tricuspid), in which the ventricles contract and allow blood to be pumped from the heart to the rest of the body through the aorta and pulmonary arteries. The second heart sound (S 2 ) occurs during ventricular diastole (closure of the sigmoid/semilunar valves: aortic and pulmonary) in which the ventricles relax and allow blood to enter from the atria.…”

Section: Introductionmentioning

confidence: 99%

“…20 If the patient has heart murmurs, the frequency spectrum could extend to 500 Hz or even higher. 6,17 These murmurs are related to the third and fourth heart sounds (S 3 and S 4 ), which are less recurrent.…”

Section: Introductionmentioning

confidence: 99%

“…8,26,28,29 For this reason, the separation between heart and lung sounds is of great interest to specialists in both areas (cardiologists and pulmonologists) since it will allow more precise diagnoses. 17,21,27,30,31 However, one of the main difficulties presented by this problem is that heart and lung sounds are not only overlapped in time, but also in frequency. 13,28 Therefore, it is necessary to develop an algorithm that allows to recover each signal while maintaining its characteristic properties, avoiding the loss of information in the process.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Source separation for single channel thoracic cardio-respiratory sounds applying Non-negative Matrix Factorization (NMF) using a focused strategy on heart sound positions

Arce¹,

Cuadra

2023

18th International Symposium on Medical Information Processing and Analysis

View full text Add to dashboard Cite

Auscultation using stethoscopes allows the diagnosis of respiratory and cardiac diseases. However, these sounds interfere with each other both in time and frequency. In the case of recording heart sounds, it is possible to ask the patient to stop their breathing to perform auscultation and obtain a pure heart sound. But, in the case of lung sounds it is impossible to do the same. In this paper, a source separation method based on Non-negative Matrix Factorization (NMF) is used to decompose a signal into different c omponents. T he m ethod proposed uses information from the estimated lung sound to reinsert the segments of interest into the original signal. The objective of this approximation is not to distort the segments of pure respiratory sound (free of heart sound). This method is compared to a base case of NMF decomposition on the raw signal. Three criteria for classifying the components based on the literature are also proposed, which will allow to indicate which component corresponds to each sound. The results were evaluated using temporal and spectral correlations, mean square error (MSE) and signal to distortion ratio (SDR) between the original respiratory signal and the respiratory signal estimated through the algorithm. It is shown that the best approximation is the NMF decomposition on the entire signal & replacing segments under different p arameter variations.

show abstract

“…The short-time Fourier transform (STFT)-based method [14] and wavelet-based filtering methods [15][16][17] have also been used by the researchers for transient removal. Most of the filtering methods follow the general procedure as the noise signal decomposition [14,[18][19][20] and then the noise reduction. WS reductions from the IS also come under these categories of transient noise reduction.…”

Section: Introductionmentioning

confidence: 99%

Hard component detection of transient noise and its removal using empirical mode decomposition and wavelet‐based predictive filter

Tanwar

Somkuwar

2018

IET signal process.

Self Cite

View full text Add to dashboard Cite

In this study, the authors propose a novel method that provides weapon sound (WS) interference reduction of military instruction sound or instructions sound (IS). This method comprises of the following steps. In the first step, the mixed signal is split into its basic constituents using principal component analysis. The second step provides the intrinsic mode functions using the empirical mode decomposition method. In the third step, the fundamental frequency component is extracted by cepstrum analysis. The localisation of WS is done using the grid search window-dominant signal subspace-based method in the fourth step. Multiscale prediction and filtering using Daubechies wavelet-based prediction are applied in the final step. The proposed method provides better results than existing baseline methods, for various signal-to-noise ratios. The simulations have also been performed on recorded WS corrupted IS signals from three different distances for the various input signal-to-weapon noise ratios, to verify the accuracy of the proposed method.

show abstract

Enhancement of lung sounds based on empirical mode decomposition and Fourier transform algorithm

Cited by 25 publications

References 37 publications

On the effectiveness of signal decomposition, feature extraction and selection on lung sound classification

On the effectiveness of signal decomposition, feature extraction and selection on lung sound classification

Source separation for single channel thoracic cardio-respiratory sounds applying Non-negative Matrix Factorization (NMF) using a focused strategy on heart sound positions

Hard component detection of transient noise and its removal using empirical mode decomposition and wavelet‐based predictive filter

Contact Info

Product

Resources

About