Respiratory sounds. Advances beyond the stethoscope

Imaging of simulated crackle sounds distribution on the chest

Crackles sounds have been associated with several pulmonary pathologies and diverse algorithms have been proposed for extracting and counting them from the acquired lung sound. These tasks depend among other factors, of the relation between the magnitude of the crackle and the background lung sound. In this work, we explore multivariate signal processing to deal with the tasks and propose a new concept, the discontinuous adventitious sounds imaging. The image formation is founded on the results of two proposed methodologies that use an autoregressive (AR) model. In the first case, the AR coefficients feed an artificial neural network (ANN) to classify temporal acoustic information as healthy or sick and; in the second case, a time-variant AR (TVAR) model, obtained by the RLS algorithm, permits to detect changes in the TVAR coefficients to be associated with the number of crackles. For AR-ANN, the ratio of the temporal windows classified as sick to the classified as healthy is used as an index to form the adventitious image, while for TVAR-RLS, an estimation of the number of crackles is obtained to form the corresponding image. The results indicated that fine and coarse crackles could be detected and counted even with very low crackle magnitude so that the formation of a crackle distribution image was consistent.

Elimination of vesicular sounds from pulmonary crackle waveforms

Pulmonary crackles and their parameters are very useful in the diagnosis of pulmonary disorders. A new automatic method has been proposed for the elimination of background vesicular sound from crackle signal with a view to introduce minimum distortion to crackle parameters. A region of interest is designated and a distortion metric based on the correlation between raw and filtered waveforms in that region is defined. Filter cut-off frequency is estimated based on the distortion metric. To reduce computational cost, a regression analysis is also realized which predicts a new fitted cut-off frequency from the estimated cut-off frequency. As a comparison basis, wavelet filtering is also applied on the same data. The algorithm is validated on simulated crackles superimposed on recorded vesicular sound with results indicating that filtering is achieved with minimal distortion of crackle parameters. The algorithm is also applied on real crackles from subjects with various respiratory disorders. The results show the extent of the effect of vesicular sound on crackle parameters, emphasizing the significance of proper filtering in crackle studies.

Reproducibility of dynamically represented acoustic lung images from healthy individuals

Background and aim:

Acoustic lung imaging offers a unique method for visualising the lung. This study was designed to demonstrate reproducibility of acoustic lung images recorded from healthy individuals at different time points and to assess intra- and inter-rater agreement in the assessment of dynamically represented acoustic lung images.

Methods:

Recordings from 29 healthy volunteers were made on three separate occasions using vibration response imaging. Reproducibility was measured using quantitative, computerised assessment of vibration energy. Dynamically represented acoustic lung images were scored by six blinded raters.

Results:

Quantitative measurement of acoustic recordings was highly reproducible with an intraclass correlation score of 0.86 (very good agreement). Intraclass correlations for inter-rater agreement and reproducibility were 0.61 (good agreement) and 0.86 (very good agreement), respectively. There was no significant difference found between the six raters at any time point. Raters ranged from 88% to 95% in their ability to identically evaluate the different features of the same image presented to them blinded on two separate occasions.

Conclusion:

Acoustic lung imaging is reproducible in healthy individuals. Graphic representation of lung images can be interpreted with a high degree of accuracy by the same and by different reviewers.

Acoustically detectable cellular-level lung injury induced by fluid mechanical stresses in microfluidic airway systems

We describe a microfabricated airway system integrated with computerized air-liquid two-phase microfluidics that enables on-chip engineering of human airway epithelia and precise reproduction of physiologic or pathologic liquid plug flows found in the respiratory system. Using this device, we demonstrate cellular-level lung injury under flow conditions that cause symptoms characteristic of a wide range of pulmonary diseases. Specifically, propagation and rupture of liquid plugs that simulate surfactant-deficient reopening of closed airways lead to significant injury of small airway epithelial cells by generating deleterious fluid mechanical stresses. We also show that the explosive pressure waves produced by plug rupture enable detection of the mechanical cellular injury as crackling sounds.

Stethoscopes: what are we hearing?

This paper develops an objective methodology to test the audio quality of stethoscopes, classifies stethoscopes into five functional categories, and compares the audio performance of each of the five categories. These categories, based on the manufacturer's recommended use, are basic assessment, cardiology, disposable, high-end cardiology, and physical assessment. The classification into categories is based on the intended performance of the stethoscopes as provided by the manufacturers. After developing the procedures and running more than 500 tests, the stethoscope with the least amount of loss over the spectrum was chosen from each of the five categories; the five were then compared to one another. Thirty-nine stethoscopes from 11 manufacturers were used in this study. The objective test methodology allows for side-by-side comparison of stethoscopes from various manufacturers that is independent of the manufacturer's published test results.

Acoustic detection of coronary artery disease

Coronary artery disease (CAD) occurs when the arteries to the heart (the coronary arteries) become blocked by deposition of plaque, depriving the heart of oxygen-bearing blood. This disease is arguably the most important fatal disease in industrialized countries, causing one-third to one-half of all deaths in persons between the ages of 35 and 64 in the United States. Despite the fact that early detection of CAD allows for successful and cost-effective treatment of the disease, only 20% of CAD cases are diagnosed prior to a heart attack. The development of a definitive, noninvasive test for detection of coronary blockages is one of the holy grails of diagnostic cardiology. One promising approach to detecting coronary blockages noninvasively is based on identifying acoustic signatures generated by turbulent blood flow through partially occluded coronary arteries. In fact, no other approach to the detection of CAD promises to be as inexpensive, simple to perform, and risk free as the acoustic-based approach. Although sounds associated with partially blocked arteries are easy to identify in more superficial vessels such as the carotids, sounds from coronary arteries are very faint and surrounded by noise such as the very loud valve sounds. To detect these very weak signals requires sophisticated signal processing techniques. This review describes the work that has been done in this area since the 1980s and discusses future directions that may fulfill the promise of the acoustic approach to detecting coronary artery disease.

Design of a DSP-based instrument for real-time classification of pulmonary sounds

Auscultation of pulmonary sounds provides valuable clinical information but has been regarded as a tool of low diagnostic value due to the inherent subjectivity in the evaluation of these sounds. In this work, a Digital Signal Processor is used to design an instrument capable of acquiring, parameterizing and subsequently classifying lung sounds into two classes with an aim to evaluate them objectively in real time. The instrument operates on sound signal from a chest microphone and flow signal from a pneumotachograph. The classification is carried out separately on the 12 reference libraries (pathological and healthy) of six sub-phases of a full respiration cycle and the results are combined to arrive at a final decision. The k-nearest neighbour and minimum distance classifiers with different distance metrics have been implemented in the instrument. The instrument was tested in the clinical environment, attaining 96% accuracy in real-time classification.

Boundary element model for simulating sound propagation and source localization within the lungs

An acoustic boundary element (BE) model is used to simulate sound propagation in the lung parenchyma. It is computationally validated and then compared with experimental studies on lung phantom models. Parametric studies quantify the effect of different model parameters on the resulting acoustic field within the lung phantoms. The BE model is then coupled with a source localization algorithm to predict the position of an acoustic source within the phantom. Experimental studies validate the BE-based source localization algorithm and show that the same algorithm does not perform as well if the BE simulation is replaced with a free field assumption that neglects reflections and standing wave patterns created within the finite-size lung phantom. The BE model and source localization procedure are then applied to actual lung geometry taken from the National Library of Medicine's Visible Human Project. These numerical studies are in agreement with the studies on simpler geometry in that use of a BE model in place of the free field assumption alters the predicted acoustic field and source localization results. This work is relevant to the development of advanced auscultatory techniques that utilize multiple noninvasive sensors to construct acoustic images of sound generation and transmission to identify pathologies.

Expiratory nasal sound analysis as a new method for evaluation of nasal obstruction in patients with nasal septal deviation: comparison of expiratory nasal sounds from both deviated and normal nasal cavity

BACKGROUND

The reliability of nasal obstruction measurements could be improved, and several new techniques are being developed. Our objective was to investigate the use of a new software program, Odiosoft-Rhino, in the assessment of nasal obstruction via analysis of the sounds of nasal expiration.

METHODS

We compared the nasal symptom scores and Odiosoft-Rhino and acoustic rhinometry test results for 61 patients with known nasal septal deviation.

RESULTS

We found a significant difference, and a correlation, between Odiosoft-Rhino results at 2000-4000 Hz and 4000-6000 Hz intervals, and the minimal cross-sectional area 2.2 cm from the nostril, in the right nasal cavity in patients with right-sided deviations. Similar results were observed for the left nasal cavity in patients with left-sided deviations.

CONCLUSIONS

The Odiosoft-Rhino software test is noninvasive, requires minimal cooperation and experience, and provides results that can be saved as digital data. Additionally, data from the Odiosoft-Rhino test are strongly correlated with acoustic rhinometry results and visual analogue scores of nasal obstruction. It seems that sound intensity within the 2000-4000 Hz and 4000-6000 Hz intervals is more sensitive than other sound intensity intervals. Thus, we speculate that Odiosoft-Rhino testing could be used as a new diagnostic method in order to evaluate nasal airflow in clinical practice.

Analysis of wheezes in asthmatic patients during spontaneous respiration

Respiratory sound analysis can offer important information related to pulmonary diseases. Wheezes have been reported as adventitious respiratory sounds in asthmatic or obstructive patients, during forced exhalation maneuvers. In this work, we propose a method for analysis of respiratory sounds in frequency domain, during spontaneous ventilation. Two databases were analyzed: signals acquired during spirometry (DBspir), composed by 23 subjects (N=15 asthmatics, N=8 control); and signals acquired during spontaneous ventilation for 120 seconds (DBsv), composed by 26 asthmatics. Using an autoregressive model (AR, order 16), it was calculated the Power Spectral Density (PSD) for each expiration and the peak frequency (fp) was estimated. Higher values of fp were found in asthmatic patients with severe obstruction in relation to light obstruction or control subjects. The effect of bronchodilator inhalation in asthmatic patients was studied in the database DBsv, analyzing contribution of wheezes in the bandwidth 600-2000 Hz (HFband)., Differences of number of respiratory cycles with wheezes (Dwheez index), before and after bronchodilator inhalation were evaluated. It was found a good correlation between Dwheez and FEV1% improvement (FEV1>%_imp), for FEV1%_imp > 10%. This method could predict the FEV1%_imp by means of estimation of Dwheez index during spontaneous ventilation.

On applying continuous wavelet transform in wheeze analysis

The identification of continuous abnormal lung sounds, like wheezes, in the total breathing cycle is of great importance in the diagnosis of obstructive airways pathologies. To this vein, the current work introduces an efficient method for the detection of wheezes, based on the time-scale representation of breath sound recordings. The employed Continuous Wavelet Transform is proven to be a valuable tool at this direction, when combined with scale-dependent thresholding. Analysis of lung sound recordings from 'wheezing' patients shows promising performance in the detection and extraction of wheezes from the background noise and reveals its potentiality for data-volume reduction in long-term wheezing screening, such as in sleep-laboratories.

Classification of lung sounds during bronchial provocation using waveform fractal dimensions

Lung sounds (LS) of children after bronchoconstriction should differ from baseline LS in terms of amplitude and pattern characteristics. To test these hypotheses, time-domain and fractal based analyses have been applied to LS acquired from eight children ages 9-15 y pre- and post-methacholine challenge (MCh). Change in forced expiratory volume in 1 s after MCh ranged from -4% to -37%, with change proportional to severity of bronchoconstriction. Sounds were recorded over the posterior right lower lung lobe while subjects breathed normally for 60 s with flow measurement, and during 10 s of breath hold (BH). From root-mean-square (RMS) of LS and BH signals, signal-to-noise ratio (SNR) was determined. Two fractal dimension (FD) algorithms were applied, based on signal variance and morphology. Feature vectors for 1-nearest-neighbor classification contained FD and RMS values within flow plateau ranges. Results for LS within 75-600 Hz indicate that the combination of RMS-SNR and morphology-based FD values offers better classification of bronchoconstriction with LS, relative to using RMS-SNR with variance-based FDs and RMS-SNR alone. True positive classification was 90.3%, 63.5% and 58.3% respectively, and false positive classification was 23.4%, 24.9% and 26.1% respectively. Both RMS-SNR and FD values provide useful insight into LS changes post-bronchoconstriction.

Dynamic visualization of lung sounds with a vibration response device: a case series

BACKGROUND

The field of computer-assisted mapping of lung sounds is constantly evolving and several devices have been developed in this field.

OBJECTIVES

Our objective was to evaluate a new computer-assisted lung sound imaging system, 'vibration response imaging' (VRI), that records and creates a dynamic image of breath sounds. We postulated that the VRI display format would qualitatively and quantitatively reveal breath sound distribution throughout the breathing cycle.

METHODS

Lung sounds were recorded from 5 healthy adults and 14 patients with various respiratory illnesses using VRI. The lung sounds were processed by the VRI software, which incorporates an algorithm to convert breath sounds in the frequency range of 150-250 Hz to a dynamic image and quantitative assessment of breath sound distribution.

RESULTS

Images and quantifications from recordings of the healthy adults showed distinct patterns for inspiration and expiration. Images and quantifications from the subjects with respiratory illness differed substantially from the images of the healthy subjects. Both healthy and pathological subjects presented some expected characteristics of breath sound distribution.

CONCLUSIONS

The VRI device may provide a new perspective in acoustic imaging and quantification of breath sounds by adding aspects of time analysis and quantification of distribution to existing methods. Further studies will be required in order to establish reliability of repeated recordings and to validate the sensitivity of the system in detecting various lung pathologies.

Using lung sounds in classification of pulmonary diseases according to respiratory subphases

Auscultation-based diagnosis of pulmonary disorders relies heavily on the presence of adventitious sounds and on the altered transmission characteristics of the chest wall. The phase information of the respiratory cycle within which adventitious sounds occur is very helpful in diagnosing different diseases. In this study, respiratory sound data belonging to four pulmonary diseases, both restrictive and obstructive, along with healthy respiratory data are used in various classification experiments. The sound data are separated into six subphases, namely, early, mid, late inspiration and expiration and classification experiments using a neural classifier are carried out for each subphase. The AR parameters acquired from segmented sound signals, prediction error and the ratio of expiration to inspiration durations are used to construct the feature set to the neural classifier. Classification experiments are carried out between healthy and pathological sound segments, between restrictive and obstructive sound segments and between two different disease sound segments. The results indicate that the classifier performance demonstrates subphase dependence for different diseases. These results may shed light in eliminating redundant feature spaces in building an expert system using lung sounds for pulmonary diagnosis.

Mobile nocturnal long-term monitoring of wheezing and cough

Changes in normal lung sounds are an important sign of pathophysiological processes in the bronchial system and lung tissue. For the diagnosis of bronchial asthma, coughing and wheezing are important symptoms that indicate the existence of obstruction. In particular, nocturnal long-term acoustic monitoring and assessment make sense for qualitative and quantitative detection and documentation. Previous methods used for lung function diagnosis require active patient cooperation that is not possible during sleep. We developed a mobile device based on the CORSA standard that allows the recording of respiratory sounds throughout the night. To date, we have recorded 133 patients with different diagnoses (80 male, 53 female), of whom 38 were children. In 68 of the patients we could detect cough events and in 87 we detected wheezing. The recording method was tolerated by all participating adults and children. Our mobile system allows non-invasive and cooperation-independent nocturnal monitoring of acoustic symptoms in the domestic environment, especially at night, when most ailments occur.

A Framework for the Analysis of Acoustical Cardiac Signals

Skilled cardiologists perform cardiac auscultation, acquiring and interpreting heart sounds, by implicitly carrying out a sequence of steps. These include discarding clinically irrelevant beats, selectively tuning in to particular frequencies and aggregating information across time to make a diagnosis. In this paper, we formalize a series of analytical stages for processing heart sounds, propose algorithms to enable computers to approximate these steps, and investigate the effectiveness of each step in extracting relevant information from actual patient data. Through such reasoning, we provide insight into the relative difficulty of the various tasks involved in the accurate interpretation of heart sounds. We also evaluate the contribution of each analytical stage in the overall assessment of patients. We expect our framework and associated software to be useful to educators wanting to teach cardiac auscultation, and to primary care physicians, who can benefit from presentation tools for computer-assisted diagnosis of cardiac disorders. Researchers may also employ the comprehensive processing provided by our framework to develop more powerful, fully automated auscultation applications.

A multi-channel device for respiratory sound data acquisition and transient detection

In this study, a multi-channel analog data acquisition and processing device with the additional feature of detecting adventitious sounds has been designed and implemented. The overall system consists of fourteen microphones attached on the backside, an airflow measuring unit, a fifteen-channel amplifier and filter unit connected to a personal computer (PC) via a data acquisition (DAQ) card, and an interface and adventitious sound detection program prepared using LabVIEW (6.0, National Instruments) and MATLAB (7.0.1, MathWorks). The system records the fourteen-channel respiratory sound data at the posterior chest wall and in addition measures the air flow to synchronize the pulmonary signal on the respiration cycle. Respiratory data are amplified and band-pass filtered, whereas flow signal is only low-pass filtered since it is a low-frequency signal with sufficiently high amplitude. All data are sent to a PC to be digitized by DAQ card, then to be processed and stored. An algorithm based on wavelet decomposition is developed which detects the adventitious pulmonary sounds, mainly the crackles and wheezes. This system is intended to be used for mapping the pulmonary sounds and detecting and locating the adventitious pulmonary sounds.

Respiratory flow estimation from tracheal sound by adaptive filters

In this study, average power of tracheal sound (Pave) was used to estimate flow by parametric method as well as adaptive filters as a nonparametric method. Based on some preliminary studies, an exponential model was used for describing the relationship between flow and Pave for parametric method. It was assumed that flow signal of at least one breath from each target flow is available for calibration. The error for flow estimation with parametric method, was found to be 9 ± 3 % and 10 ± 4 % for inspiration and expiration, respectively. Considering nonparametric method, the estimation error was the least for the third order adaptive filter using the average power of the tracheal sound (dB), which was 10 ± 3 % and 11 ± 4 % for inspiration and expiration, respectively.

Transmission of lung sounds through light clothing

BACKGROUND

Doctors are exhorted to always place the stethoscope directly on the skin and never to auscultate through clothing. Nevertheless, casual observation reveals that doctors and even pulmonologists often violate this principle.

OBJECTIVES

This study was designed to evaluate the sensitivity of two common stethoscopes when used through clothing.

METHODS

Littmann Classic and Littmann Master Cardiology stethoscopes were studied under conditions of light (60-100 g), medium (240 g) and heavy (555 g) force when placed on a lung sound test platform with one or two layers of cloth (T-shirt material and flannel) interposed between the stethoscope and the test surface. The test platform was designed to mimic the acoustic and mechanical properties of the chest wall and was driven by amplified white noise. The recorded amplitude spectra were compared over a range of 150-1,000 Hz.

RESULTS

Compared to the sensitivity on a bare test platform surface, either fabric in single or double layers attenuated the sounds by a mean of 5-18 dB under light pressure. This attenuation was nearly abolished by the addition of either medium or heavy force on the stethoscope head.

CONCLUSIONS

The deleterious effect of one or two layers of indoor clothing on lung sounds acquired through a stethoscope can be negated by force on the stethoscope head making effective auscultation possible. Nevertheless, auscultation through clothing remains problematic due to the hindrance to inspection and percussion and the risk of acoustic artifacts caused by clothing.

Physical diagnosis of chronic obstructive pulmonary disease

Among the various diagnostic strategies of chronic obstructive pulmonary disease (COPD), physical diagnosis is the quickest and requires no extra cost. Rapid physical diagnosis of COPD in primary care practice can lead to earlier actions of preventive measures and counseling for patients. Further, rapid physical diagnosis of COPD in an emergency department is also crucial for timely use of potentially lifesaving therapy specific for COPD patients. In this review, we will present a broad scope of physical findings for rapid physical diagnosis of COPD.

Sensitivity of pulmonary crackle parameters to filter cut-off frequency

Pulmonary crackles are very important indicators in the diagnosis of pulmonary disorders. Parameters derived from their waveforms, such as initial deflection width and two-cycle duration width, are useful in classifying a crackle and correlating it with various disorders. To obtain a crackle waveform, the background vesicular sound is filtered out and the crackle parameters are very sensitive to filter cut-off frequency. In this study, sensitivity analysis of waveform parameters of simulated crackles superimposed on real vesicular sound is carried out, emphasizing the necessity of precision in determining the filter cut-off frequency.

Wheeze detection based on time-frequency analysis of breath sounds

Abnormal breath sounds like wheezes are observed in patients with obstructive pulmonary diseases. The aim of this study was to construct an automatic technique for wheeze detection and monitoring using spectral analysis. Wheezes from 13 patients with diagnosed asthma, chronic obstructive pulmonary disease and pneumonia were recorded and a time-frequency wheeze detector (TF-WD) based on TF wheeze characteristics was constructed. The TF-WD was evaluated using 337 wheezes by comparing its findings with those from clinical auscultation performed by two experts. In addition, the TF-WD was tested against artificial noise. The experimental and testing results justified the efficient performance and high noise robustness of the TF-WD.

Terminologia da ausculta pulmonar utilizada em publicações médicas brasileiras, no período de janeiro de 1980 a dezembro de 2003

OBJETIVO: Avaliar a adequação de uso de termos semiológicos da ausculta pulmonar em publicações médicas brasileiras sobre doenças respiratórias, no período de janeiro de 1980 a dezembro de 2003. MÉTODOS: Realizou-se um estudo descritivo, analisando-se três revistas médicas: Jornal de Pneumologia, Jornal de Pediatria e Revista Médica Brasileira. Foram selecionados os artigos originais e relatos de casos sobre doenças respiratórias, de onde foram extraídos os termos semiológicos da ausculta pulmonar. Foi avaliada a adequação dos termos na descrição dos ruídos adventícios. RESULTADOS: Encontrou-se maior inadequação no uso dos termos de ruídos descontínuos, comparado com o uso dos termos de ruídos contínuos (87,7% versus 44%, p = 0,0000). Não houve diferença significativa entre relatos de pneumologistas e de outros especialistas quanto à inadequação no uso dos termos (56,5% versus 62,0%, p = 0,26). Também não observamos diferença significativa entre as regiões do país e os períodos antes e após a divulgação da nomenclatura internacional. CONCLUSÃO: O uso inadequado dos termos para descrever ruídos adventícios na ausculta pulmonar continua sendo um fenômeno freqüente e geral nas publicações médicas brasileiras.

Nonlinear analysis of wheezes using wavelet bicoherence

Wheezes, as being abnormal breath sounds, are observed in patients with obstructive pulmonary diseases, such as asthma. The aim of this study was to capture and analyze the nonlinear characteristics of asthmatic wheezes, reflected in the quadrature phase coupling of their harmonics, as they evolve over time within the breathing cycle. To achieve this, the continuous wavelet transform (CWT) was combined with third-order statistics/spectra. Wheezes from patients with diagnosed asthma were drawn from a lung sound database and analyzed in the time-bi-frequency domain. The analysis results justified the efficient performance of this combinatory approach to reveal and quantify the evolution of the nonlinearities of wheezes with time.

Design, construction, and evaluation of a bioacoustic transducer testing (BATT) system for respiratory sounds

Many different transducers are employed for recording respiratory sounds including accelerometers and microphones in couplers. However, there is no standard lung sound transducer or any device to compare transducers so that measurements from different laboratories can be determined to be of physiologic origin rather than technical artifacts of the transducers. To address this problem, we designed and constructed a prototype of a device that can be used to compare accelerometers, microphones enclosed in couplers, and stethoscopes. The prototype device consists of a rigid chamber containing a loudspeaker that opens to an antechamber covered by a viscoelastic material with mechanical properties similar to human skin and subcutaneous tissue. When driven by a white noise source, we found the sound output at the surface to be useful to comparatively evaluate sensors between 100 and 1200 Hz where lung sounds have most of their spectral energy. We compared the viscoelastic layer to similar thicknesses of fresh meat and fat and found them to produce similar acoustic spectra. This device allows air-coupled transducers, accelerometers, and stethoscopes used in respiratory sounds measurements to be compared under physical conditions similar to their intended use.

Monte Carlo simulation of liquid bridge rupture: application to lung physiology

In the course of certain lung diseases, the surface properties and the amount of fluids coating the airways changes and liquid bridges may form in the small airways blocking the flow of air, impairing gas exchange. During inhalation, these liquid bridges may rupture due to mechanical instability and emit a discrete sound event called pulmonary crackle, which can be heard using a simple stethoscope. We hypothesize that this sound is a result of the acoustical release of energy that had been stored in the surface of liquid bridges prior to its rupture. We develop a lattice gas model capable of describing these phenomena. As a step toward modeling this process, we address a simpler but related problem, that of a liquid bridge between two planar surfaces. This problem has been analytically solved and we use this solution as a validation of the lattice gas model of the liquid bridge rupture. Specifically, we determine the surface free energy and critical stability conditions in a system containing a liquid bridge of volume Omega formed between two parallel planes, separated by a distance 2h, with a contact angle Theta using both Monte Carlo simulation of a lattice gas model and variational calculus based on minimization of the surface area with the volume and the contact angle constraints. In order to simulate systems with different contact angles, we vary the parameters between the constitutive elements of the lattice gas. We numerically and analytically determine the phase diagram of the system as a function of the dimensionless parameters hOmega(-1/3) and Theta. The regions of this phase diagram correspond to the mechanical stability and thermodynamical stability of the liquid bridge. We also determine the conditions for the symmetrical versus asymmetrical rupture of the bridge. We numerically and analytically compute the release of free energy during rupture. The simulation results are in agreement with the analytical solution. Furthermore, we discuss the results in connection to the rupture of similar bridges that exist in diseased lungs.

A fuzzy expert system design for analysis of body sounds and design of an unique electronic stethoscope (development of HILSA kit)

In this paper we have developed a fuzzy expert system (FES) for different sounds produced by different organs in the human body. We have also constructed a unique electronic stethoscope. The human body sounds produced by different organs like heart, lungs and intestine were analyzed. The doctor provided the data and relation between variables chosen for each organ sound. Using this information a rule base for fuzzy expert system was built. Such FES helps the medical doctor in arriving at appropriate decision in different difficult clinical situations. The examination of body sounds was done using conventional stethoscope (CS) and electronic stethoscope (ES), which was uniquely designed for this study. We have found that unique stethoscope developed by us is far superior to conventional stethoscope by its overall performance.

Acoustic effects of positive end-expiratory pressure on normal lung sounds in mechanically ventilated pigs

Computerized lung sounds analysis offers a new technique to monitor regional ventilation during spontaneous breathing. The purpose of the present study was to assess the acoustic behaviour of the respiratory system in healthy pigs during mechanical ventilation when a positive end-expiratory pressure (PEEP) is applied. Lung sounds were recorded during mechanical ventilation and different PEEP levels of 0, 5, 10, 15 and 20 cm H(2)O were applied. The increase in end-expiratory lung volume (EELV) related to the PEEP application was also measured and the correlation between changes in EELV (DeltaEELV) and sound amplitude (DeltaA) was examined. The amplitude of normal lung sounds was reduced by application of PEEP >or=10 cm H(2)O (P<0.05). The increase in PEEP from 0 to 20 cm H(2)O reduced the acoustic energy of lung sounds recorded at ZEEP by 0.3 dB (PEEP 5), 2 dB (PEEP 10), 5 dB (PEEP 15) and 7 dB (PEEP 20), which corresponds to 1%, 6%, 14% and 21% in acoustic attenuation, respectively. The variations in DeltaA correlated with changes in lung volume (P<0.05) and with changes in compliance of the respiratory system (P<0.05), but were not correlated with changes of the resistance of respiratory system. The frequency analysis showed a downward shifting of the spectra at frequencies between 150 and 600 Hz for PEEP levels >or=10 cm H(2)O and frequencies between 75 and 600 Hz for PEEP levels >or=15 cm H(2)O. The application of increasing levels of PEEP reduced the amplitude and changed the spectral characteristics of normal lung sounds.

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.

Multi-channel classification of respiratory sounds

In this study, respiratory sounds of pathological and healthy subjects were analyzed via frequency spectrum and AR model parameters with a view to construct a diagnostic aid based on auscultation. Each subject is represented by 14 channels of respiratory sound data of a single respiration cycle. Two reference libraries, pathological and healthy, were built based on multi-channel respiratory sound data for each channel and for each respiration phase, inspiration and expiration, separately. A multi-channel classification algorithm using K nearest neighbor (k-NN) classification method was designed. Performances of the two classifiers using spectral feature set corresponding to quantile frequencies and 6th order AR model coefficients on inspiration and expiration phases are compared.

Heart sounds interference cancellation in lung sounds

Several attempts have been made to achieve a quantitative analysis of lung sounds mainly for two purposes: a) an understanding of their genesis, and b) an insight of their changes with pathologies for medical diagnosis. Early studies involved the collection of acoustic information at several positions on the thoracic surface or at the extra-thoracic trachea with one up to four microphones, but with a non-simultaneous acquisition. However, an increment for simultaneous acquisition points has been suggested; for example, as a consequence of multichannel acquisition acoustic visualization through computerized interpolation has emerged being helpful to analyze lung sounds (LS) origin, distribution, and relation to ventilation. Nevertheless, quantitative analysis of lung sounds requires eliminating interference signals prior to the extraction of relevant features. The acquired signals not only contain LS but also heart sounds (HS) among other interferences. HS are unavoidable and sometimes represent severe disturbing interference. This paper proposes a HS cancellation scheme as an extension of a previous effort using the Empirical Mode Decomposition (EMD) and a combination of time warping with linear adaptive FIR filtering. Simulated signals are used to evaluate the performance of the proposed scheme under known and controlled scenarios.

Qualitative and quantitative evaluation of heart sound reduction from lung sound recordings

Recursive least squares (RLS) adaptive noise cancellation (ANC) and wavelet transform (WT) ANC have been applied and compared for heart sound (HS) reduction from lung sounds (LS) recordings. Novel processes for quantitative and qualitative evaluation of any method for HS reduction from LS have also been proposed.

Combining neural network and genetic algorithm for prediction of lung sounds

Recognition of lung sounds is an important goal in pulmonary medicine. In this work, we present a study for neural networks-genetic algorithm approach intended to aid in lung sound classification. Lung sound was captured from the chest wall of The subjects with different pulmonary diseases and also from the healthy subjects. Sound intervals with duration of 15-20 s were sampled from subjects. From each interval, full breath cycles were selected. Of each selected breath cycle, a 256-point Fourier Power Spectrum Density (PSD) was calculated. Total of 129 data values calculated by the spectral analysis are selected by genetic algorithm and applied to neural network. Multilayer perceptron (MLP) neural network employing backpropagation training algorithm was used to predict the presence or absence of adventitious sounds (wheeze and crackle). We used genetic algorithms to search for optimal structure and training parameters of neural network for a better predicting of lung sounds. This application resulted in designing of optimum network structure and, hence reducing the processing load and time.

Response to bronchodilator in infants with bronchiolitis can be predicted from wheeze characteristics

OBJECTIVE

Lung sounds analysis has been used for clinical care. Our objectives were to characterize the spectral pattern of lung sounds and their relation to bronchodilator effects in acute bronchiolitis (AB). We hypothesized that patients with sinusoidal wheezes (SW) would show a more significant bronchodilator response.

METHODOLOGY

We studied 22 asleep hospitalized infants (14 boys, eight girls), aged 5.2 +/- 1 months, 16 with a positive respiratory syncytial virus test, during their first 3 days after admission. Patients breathed spontaneously through a face mask connected to a pneumotachograph during normal breathing, and only target flows of 0.1 +/- 0.02 L/s were analyzed. Sounds were obtained using two contact sensors attached over both posterior lower lobes. For inspiratory and expiratory sounds, we determined the frequencies below which 25% (F25), 50% (F50), 75% (F75) and 99% (SEF99) of the spectral power between 100 and 1000 Hz was contained. We repeated the measurements 20 min after bronchodilator therapy in all patients.

RESULTS

We found classic SW in 11 patients, while the other 11 had complex wheezes (CW). There were positive bronchodilator responses in 9/11 with SW and 3/11 with CW (P < 0.01). Patients who responded to salbutamol showed an increase in power at low frequencies after medication (P < 0.01), and a positive correlation between wheezing and the increase in the power spectra measured by F50 and SEF99 (P < 0.001).

CONCLUSIONS

We conclude that sinusoidal and complex wheezes occur in patients with AB, that a positive response to bronchodilator is significantly more common in those with classic SW and that lung sounds analysis is a reproducible, safe and non-invasive method for assessing wheeze in infants.

Response of acoustic transmission to positive airway pressure therapy in experimental lung injury

OBJECTIVE

To evaluate the effect of positive end-expiratory pressure on the sound filtering characteristics of injured lungs.

DESIGN AND SETTING

Prospective experimental study in the animal laboratory in an academic medical center.

PATIENTS AND PARTICIPANTS

Six 35- to 45-kg anesthetized, intubated pigs.

INTERVENTIONS

Acute lung injury with intravenous oleic acid.

MEASUREMENTS AND RESULTS

We injected a multifrequency broad-band sound signal into the airway while recording transmitted sound at three locations bilaterally on the chest wall. Oleic acid injections effected a severe pulmonary edema predominantly in the dependent lung regions, with an average increase in venous admixture from 16+/-14% to 57+/-13% and a reduction in static respiratory system compliance from 31+/-6 to 16+/-3 ml/cm H(2)O. A significant concomitant increase in sound transfer function amplitude was seen in the dependent and lateral lung regions; little change occurred in the nondependent areas. The application of PEEP resulted in a decrease in venous admixture, increase in respiratory system compliance, and return of the sound transmission to preinjury levels.

CONCLUSIONS

Acute lung injury causes regional acoustic transmission abnormalities that are reversed during alveolar recruitment with PEEP.

Automatic wheeze detection based on auditory modelling

Automatic wheeze detection has several potential benefits compared with reliance on human auscultation: it is experience independent, an automated historical record can easily be kept, and it allows quantification of wheeze severity. Previous attempts to detect wheezes automatically have had partial success but have not been reliable enough to become widely accepted as a useful tool. In this paper an improved algorithm for automatic wheeze detection based on auditory modelling is developed, called the frequency- and duration-dependent threshold algorithm. The mean frequency and duration of each wheeze component are obtained automatically. The detected wheezes are marked on a spectrogram. In the new algorithm, the concept of a frequency- and duration-dependent threshold for wheeze detection is introduced. Another departure from previous work is that the threshold is based not on global power but on power corresponding to a particular frequency range. The algorithm has been tested on 36 subjects, 11 of whom exhibited characteristics of wheeze. The results show a marked improvement in the accuracy of wheeze detection when compared with previous algorithms.

Can airway obstruction be estimated by lung auscultation in an emergency room setting?

OBJECTIVE

Lung auscultation is a central part of the physical examination at hospital admission. In this study, the physicians' estimation of airway obstruction by auscultation was determined and compared with the degree of airway obstruction as measured by FEV(1)/FVC values.

METHODS

Two hundred and thirty-three patients consecutively admitted to the medical emergency room with chest problems were included. After taking their history, patients were auscultated by an Internal Medicine registrar. The degree of airway obstruction had to be estimated (0=no, 1=mild, 2=moderate and 3=severe obstructed) and then spirometry was performed. Airway obstruction was defined as a ratio of FEV(1)/FVC <70%. The degree of airway obstruction was defined on FEV(1)/FVC as mild (FEV(1)/FVC <70% and >50%), moderate (FEV(1)/FVC <50% >30%) and severe (FEV(1)/FVC <30%).

RESULTS

One hundred and thirty-five patients (57.9%) had no sign of airway obstruction (FEV(1)/FVC >70%). Spirometry showed a mild obstruction in 51 patients (21.9%), a moderate obstruction in 27 patients (11.6%) and a severe obstruction in 20 patients (8.6%). There was a weak but significant correlation between FEV(1)/FVC and the auscultation-based estimation of airway obstruction in Internal Medicine Registrars (Spearman's rho=0.328; P<0.001). The sensitivity to detect airway obstruction by lung auscultation was 72.6% and the specificity only 46.3%. Thus, the negative predictive value was 68% and the positive predictive value 51%. In 27 patients (9.7%), airway obstruction was missed by lung auscultation. In these 27 cases, the severity of airway obstruction was mild in 20 patients, moderate in 5 patients and severe in 2 patients. In 82 patients (29.4%) with no sign of airway obstruction (FEV(1)/FVC >70%), airway obstruction was wrongly estimated as mild in 42 patients, as moderate in 34 patients and as severe in 6 patients, respectively. By performing multiple logistic regression, normal lung auscultation was a significant and independent predictor for not having an airway obstruction (OR 2.48 (1.43-4.28); P=0.001).

CONCLUSION

Under emergency room conditions, physicians can quite accurately exclude airway obstruction by auscultation. Normal lung auscultation is an independent predictor for not having an airway obstruction. However, airway obstruction is often overestimated by auscultation; thus, spirometry should be performed.

Two-stage classification of respiratory sound patterns

The classification problem of respiratory sound signals has been addressed by taking into account their cyclic nature, and a novel hierarchical decision fusion scheme based on the cooperation of classifiers has been developed. Respiratory signals from three different classes are partitioned into segments, which are later joined to form six different phases of the respiration cycle. Multilayer perceptron classifiers classify the parameterized segments from each phase and decision vectors obtained from different phases are combined using a nonlinear decision combination function to form a final decision on each subject. Furthermore a new regularization scheme is applied to the data to stabilize training and consultation.

Neural classification of lung sounds using wavelet coefficients

Electronic auscultation is an efficient technique to evaluate the condition of respiratory system using lung sounds. As lung sound signals are non-stationary, the conventional method of frequency analysis is not highly successful in diagnostic classification. This paper deals with a novel method of analysis of lung sound signals using wavelet transform, and classification using artificial neural network (ANN). Lung sound signals were decomposed into the frequency subbands using wavelet transform and a set of statistical features was extracted from the subbands to represent the distribution of wavelet coefficients. An ANN based system, trained using the resilient back propagation algorithm, was implemented to classify the lung sounds to one of the six categories: normal, wheeze, crackle, squawk, stridor, or rhonchus.

Detection of Respiratory Sounds at the External Ear

Several clinical and ambulatory settings necessitate respiratory monitoring without a mouthpiece or facemask. Several studies have demonstrated the utility of breathing sound measurements performed on the chest or neck to detect airflow. However, there are limitations to skin surface measurements, including susceptibility to external noise and transducer motion. Thus, this two-part study investigated a novel location for breathing sound measurements: the external ear. The first study investigated characteristics of sound transmission from the oropharynx to the external ear in 19 adults (nine males). Broadband noise was directed into the oropharynx through a tube and mouthpiece and measured indirectly via an accelerometer affixed to the cheek. Resultant transmission to the external ear was measured with a microphone inserted into an earplug that provided acoustic isolation from ambient noise. Near-unity coherence estimates (> 0.9) between the sounds recorded at the external ear and the oropharynx were observed up to approximately 800 Hz, indicating a low-frequency region of preferred transmission. In the second study, each of 20 subjects (nine males) breathed through a pneumotachograph at targeted shallow (3.0 mL/s/kg) and tidal (7.5 mL/s/kg) flows normalized to body mass, and the resulting sounds were recorded at the external ear. Recordings during breath hold measured background noise. Shallow and tidal expiratory flows, respectively, produced signal-plus-noise-to-noise [(S + N)/N] ratios of 6.7 +/- 4.1 dB and 14.0 +/- 5.3 dB (mean +/- standard deviation) across all subjects between 150 and 300 Hz. Concurrent inspiration demonstrated (S + N)/N ratios of 6.6 +/- 3.9 dB and 14.9 +/- 6.3 dB. Thus, the external ear shows promise as an anatomic site to detect and monitor breathing in a relatively noninvasive and unobtrusive manner.