Heart Sound Cancellation Research Articles

Performance evaluation of heart sound cancellation in FPGA hardware implementation for electronic stethoscope.

This paper presents the design and evaluation of the hardware circuit for electronic stethoscopes with heart sound cancellation capabilities using field programmable gate arrays (FPGAs). The adaptive line enhancer (ALE) was adopted as the filtering methodology to reduce heart sound attributes from the breath sounds obtained via the electronic stethoscope pickup. FPGAs were utilized to implement the ALE functions in hardware to achieve near real-time breath sound processing. We believe that such an implementation is unprecedented and crucial toward a truly useful, standalone medical device in outpatient clinic settings. The implementation evaluation with one Altera cyclone II–EP2C70F89 shows that the proposed ALE used 45% resources of the chip. Experiments with the proposed prototype were made using DE2-70 emulation board with recorded body signals obtained from online medical archives. Clear suppressions were observed in our experiments from both the frequency domain and time domain perspectives.

Introduction to Special Issue

The history of chest auscultation by unaided ears goes back to almost 2000 years ago. But the discovery and use of respiratory lung sounds as indicators of health and disease began with Laennec, who formalized the relationship between human pulmonary diseases and respiratory auscultation in 1819. He then invented stethoscope in 1921. Stethoscope enabled physicians to listen to respiratory sounds of their patients and detect any symptomatic signs. Modern technology can help physicians even more these days. Breath sound analysis using digital signal processing techniques has recently drawn much attention because of its diagnostic capabilities. The articles in this special issue focus on how computerized respiratory sound analysis can quantify changes in lung sounds, store records of the measurements made, de-noise the signals of interest from any artifacts and interference, and produce graphical representations of characteristic features of the respiratory sounds to help with the diagnosis and treatment of patients suffering from lung diseases. Ray Murphy reviews the combined use of stethoscope and computerized lung sound analysis in the clinical assessment of lung infectious diseases in the first paper published in this volume. One issue of interest for both physicians and researchers in this field is to find features which best describe the relationship between the respiratory flow and the lung sounds of patients compared to that of healthy individuals. Since lung sounds have relatively low frequency and low intensity, it is crucial to remove the noise and other interfering sounds (i.e., heart sounds) from the lung sounds prior to any diagnostic analysis. Many researchers around the world have been involved in this research applying linear and nonlinear digital signal processing techniques to develop novel methods that remove adventitious sounds form the lung sound record. The paper by Gnitecki and Moussavi reviews a number of heart sound cancellation methods, and the authors discuss some advantages and problems associated with different methods. Adventitious lung sounds that are of diagnostic value include crackles, wheezes, squawks, rhonchi, and stridor. Out of these sounds, detection of crackles (both fine and coarse) is more important because of its diagnostic values at the early stages of disease. Nowadays specialists rely on computer aided analysis to count the number of crackles at different locations of chest. Hence, this area of research is concerned with developing more accurate and efficient methods for detecting crackles. The next two papers by Hadjileontiadis and Charleston et al. present two methods for crackle detection using empirical mode decomposition and fractal dimension analysis.

Heart sound cancellation from lung sound record using cyclostationarity

From the mechanism of heart sound generation, it is known that heart sounds are cyclic following the frequency of the heartbeat. This paper proposes a short-time cyclic frequency spectrum to calculate the instantaneous cycle frequency (ICF) of heart sounds as an estimation of the frequency of the heartbeat. Heart sounds in a lung sound record are detected with the assistance of ICF. Lung sounds (LSs) are recovered by removing heart sounds from the LS record. An LS record is the only input signal source; no other reference signal is necessary. Evaluation by visual inspection, auditory listening and spectral analysis all show that heart sounds are successfully cancelled without hampering the main components of lung sounds.

Localizing Heart Sounds in Respiratory Signals Using Singular Spectrum Analysis

Respiratory sounds are always contaminated by heart sound interference. An essential preprocessing step in some of the heart sound cancellation methods is localizing primary heart sound components. Singular spectrum analysis (SSA), a powerful time series analysis technique, is used in this paper. Despite the frequency overlap of the heart and lung sound components, two different trends in the eigenvalue spectra are recognizable, which leads to find a subspace that contains more information about the underlying heart sound. Artificially mixed and real respiratory signals are used for evaluating the performance of the method. Selecting the appropriate length for the SSA window results in good decomposition quality and low computational cost for the algorithm. The results of the proposed method are compared with those of well-established methods, which use the wavelet transform and entropy of the signal to detect the heart sound components. The proposed method outperforms the wavelet-based method in terms of false detection and also correlation with the underlying heart sounds. Performance of the proposed method is slightly better than that of the entropy-based method. Moreover, the execution time of the former is significantly lower than that of the latter.

Heart Sound Cancellation Based on Multiscale Products and Linear Prediction

This paper presents a novel method for Heart Sound (HS) cancellation from Lung Sound (LS) records. The method uses the multiscale product of the wavelet coefficients of the original signal to detect HS-included segments. Once the HS segments are identified, the method removes them from the wavelet coefficients at every level and estimates the created gaps by using a set of linear prediction filters. It is shown that if the segment to be predicted is stationary, a final record with no audible artifacts such as clicks can be reconstructed using this approach. The results were promising for HS removal from LS records and showed no hampering of the main components of the LS. The results were confirmed both qualitatively by listening to the reconstructed signal and quantitatively by spectral analysis.

Heart sound cancellation from lung sound recordings using time-frequency filtering

During lung sound recordings, heart sounds (HS) interfere with clinical interpretation of lung sounds over the low frequency components which is significant especially at low flow rates. Hence, it is desirable to cancel the effect of HS on lung sound records. In this paper, a novel HS cancellation method is presented. This method first localizes HS segments using multiresolution decomposition of the wavelet transform coefficients, then removes those segments from the original lung sound record and estimates the missing data via a 2D interpolation in the time-frequency (TF) domain. Finally, the signal is reconstructed into the time domain. To evaluate the efficiency of the TF filtering, the average power spectral density (PSD) of the original lung sound segments with and without HS over four frequency bands from 20 to 300 Hz were calculated and compared with the average PSD of the filtered signals. Statistical tests show that there is no significant difference between the average PSD of the HS-free original lung sounds and the TF-filtered signal for all frequency bands at both low and medium flow rates. It was found that the proposed method successfully removes HS from lung sound signals while preserving the original fundamental components of the lung sounds.

Heart sound cancellation from lung sound recordings using recurrence time statistics and nonlinear prediction

Heart sounds (HS) obscure the interpretation of lung sounds (LS). This letter presents a new method to detect and remove this undesired disturbance. The HS detection algorithm is based on a recurrence time statistic that is sensitive to changes in a reconstructed state space. Signal segments that are found to contain HS are removed, and the arising missing parts are replaced with predicted LS using a nonlinear prediction scheme. The prediction operates in the reconstructed state space and uses an iterated integrated nearest trajectory algorithm. The HS detection algorithm detects HS with an error rate of 4% false positives and 8% false negatives. The spectral difference between the reconstructed LS signal and an LS signal with removed HS was 0.34/spl plusmn/0.25, 0.50/spl plusmn/0.33, 0.46/spl plusmn/0.35, and 0.94/spl plusmn/0.64 dB/Hz in the frequency bands 20-40, 40-70, 70-150, and 150-300 Hz, respectively. The cross-correlation index was found to be 99.7%, indicating excellent similarity between actual LS and predicted LS. Listening tests performed by a skilled physician showed high-quality auditory results.

Reproducibility of the Vesicular Breath Sounds in Normal Subjects

Nonfiltered (NF) lung sounds from the apical area of the heart along with lung volumes and ECG signals were recorded from 5 normal subjects. The signals were digitized and subjected to three methods of heart sound cancellation: 75-Hz high-pass filtering (75 HF), ECG-triggered blanking (BL) and adaptive noise cancelling (AF) [IEEE Trans. Biomed. Engng 33: 1141-1148, 1986]. The sound signals were then subjected to the fast Fourier transform algorithm to obtain power spectra. Five breaths from each subject were analyzed, and their spectra were similar and slightly skewed to the right. The average values of mean, median and mode frequencies of the whole breath of 5 subjects, respectively, were for NF: 64.62 +/- 3.74, 44.57 +/- 2.06 and 36.75 +/- 1.79 Hz; for 75 HF: 150.42 +/- 17.49, 114.02 +/- 6.43 and 86.16 +/- 3.13 Hz; for BL: 81.76 +/- 6.02, 52.36 +/- 2.79, 41.10 +/- 3.15 Hz; for AF: 96.87 +/- 11.58, 68.23 +/- 10.44 and 52.25 +/- 8.97 Hz. These values showed no differences between subjects. The F values obtained by the two-way analysis of variance of all breaths of all subjects (mean, median, mode) were: NF: 0.161, 0.341, 0.089; 75 HF: 0.455, 0.042, 0.085; BL: 0.108, 0.082, 0.057; AF: 0.130, 0.204, 0.113 (all p greater than 0.1). The data revealed a remarkable lack of variation within and between subjects, suggesting similar sites and mechanisms of production and transmission.

Inspiratory and Expiratory Vesicular Breath Sounds

Unfiltered breath sounds (NF) from the apical area of the heart, lung volume and ECG signals were recorded in 5 normal subjects. The signals were digitized and subjected to three methods of heart sound cancellation: 75-Hz high-pass filtering (75 HF), ECG-triggered blanking (BL) and adaptive filtering (AF). The sound signals were then subjected to the fast Fourier transform algorithm to obtain power spectra. Inspiratory and expiratory phase sounds of five breaths of each subject were analyzed separately. The inspiratory and expiratory sound power spectra were very similar and skewed slightly to the right, and therefore characterized by median frequencies. The differences between inspiratory and expiratory median frequencies were insignificant for NF: 42.90 +/- 2.03 (mean +/- SD) vs. 46.64 +/- 2.53 Hz (p greater than 0.1); for 75 HF: 106.43 +/- 10.27 vs. 118.22 +/- 6.30 Hz (p greater than 0.5); for BL: 44.46 +/- 3.33 vs. 66.73 +/- 2.93 Hz (p greater than 0.1), for AF: 49.72 +/- 5.68 vs. 79.20 +/- 13.07 Hz (p greater than 0.1). We conclude that the lack of significant differences suggests similar mechanisms and sites of production of inspiratory and expiratory vesicular breath sounds.