Journal of Applied Physiology Information on EB 2010
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

J Appl Physiol 101: 469-476, 2006. First published April 20, 2006; doi:10.1152/japplphysiol.00273.2006
8750-7587/06 $8.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow All Versions of this Article:
101/2/469    most recent
00273.2006v1
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Web of Science (4)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kraman, S. S.
Right arrow Articles by Pasterkamp, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kraman, S. S.
Right arrow Articles by Pasterkamp, H.

Comparison of lung sound transducers using a bioacoustic transducer testing system

Steve S. Kraman,1 George R. Wodicka,2 Gary A. Pressler,2 and Hans Pasterkamp3

1Department of Internal Medicine, University of Kentucky, Lexington, Kentucky; 2Weldon School of Biomedical Engineering, Purdue University, West Lafayette, Indiana; and 3Department of Pediatrics and Child Health, University of Manitoba, Winnipeg, Manitoba, Canada

Submitted 2 March 2006 ; accepted in final form 10 April 2006


   ABSTRACT
TOP
 ABSTRACT
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Sensors used for lung sound research are generally designed by the investigators or adapted from devices used in related fields. Their relative characteristics have never been defined. We employed an artificial chest wall with a viscoelastic surface and a white noise signal generator as a stable source of sound to compare the frequency response and pulse waveform reproduction of a selection of devices used for lung sound research. We used spectral estimation techniques to determine frequency response and cross-correlation of pulses to determine pulse shape fidelity. The sensors evaluated were the Siemens EMT 25 C accelerometer (Siemens); PPG 201 accelerometer (PPG); Sony ECM-T150 electret condenser microphone with air coupler (air coupler; with cylindrical air chambers of 5-, 10-, and 15-mm diameter and conical air chamber of 10-mm diameter); Littman classic stethoscope head (Littman) connected to an electret condenser microphone; and the Andries Tek (Andries) electronic stethoscope. We found that the size and shape of the air coupler chamber to have no important effect on the detected sound. The Siemens, air coupler, and Littman performed similarly with relatively flat frequency responses from 200 to 1,200 Hz. The PPG had the broadest frequency response, with useful sensitivity extending to 4,000 Hz. The Andries' frequency response was the poorest above 1,000 Hz. Accuracy in reproducing pulses roughly corresponded with the high-frequency sensitivity of the sensors. We conclude that there are important differences among commonly used lung sound sensors that have to be defined to allow the comparison of data from different laboratories.

respiratory sounds; respiratory acoustics; microphones; accelerometers

THE ABILITY TO ELECTRONICALLY record and analyze biological sounds first developed in the 1950s with the studies by McKusick et al. (22–24) of heart and respiratory sounds, and it was soon adopted by others in the 1960s and 1970s (8, 12, 13, 28, 34, 43, 49). Since that time, it has been possible to describe lung sounds in objective terms of their timing in the respiratory cycle, duration, waveform, and frequency components. The names used for different respiratory sounds, many in common usage since Leannec's time, took on definitions based in part on their objective features revealed by sound analysis (21).

Lung sounds are frequently recorded for teaching purposes and analyzed for research. The implicit goal of such research is to extract from the lung sound signal, qualitative or quantitative information that relates to important physiological or pathological processes. The fact that these sounds may be easily and noninvasively recorded adds to the attraction of the approach.

Despite the research of the past few decades, there has been little commercial interest in lung sound analysis, and consequently no large-scale development and validation of specialized lung sound sensors has taken place. Investigators either built their own equipment or adapted sensors that had been designed for other purposes. For this reason, comparisons of data from different laboratories have been difficult. Although the different performance of sensors is of little importance in determining timing of lung sounds, it becomes a real problem when dealing with spectral analysis or the shape of waveforms because the type of sensor and means of attachment can potentially affect these features.

In the absence of a standard sensor, a standard sound source could be used to characterize sensors so their similarities and differences would be known. Until recently, such a source has not been devised, and the best that has been tried is to use human subjects, breathing under tightly controlled conditions, as sound sources (9, 19, 31, 45). Such a human lung sound standard is hardly practical or credible because of the normal variability in lung sound qualities among different people and at different locations on the chest. It was to address this deficiency that we recently designed and constructed a test platform intended to characterize the performance of commonly used lung sound transducers (18).

Our intent in designing this device was to emulate, as much as possible, the conditions under which lung sound transducers are used but employing a well-defined, stable, and reproducible sound source. Both accelerometers and air-coupled microphones have been commonly used for recording lung sounds, and therefore the testing device had to be suitable for both. Accelerometers are used in industry for the measurement of vibration on surfaces. Specially designed shaker tables are used to determine their sensitivity and frequency response. An air-coupled microphone is a microphone element placed within a coupler that contains a chamber sealed to the skin, commonly by a double-sided tape ring or strap. Skin vibrations produce air pressure fluctuations within the chamber cavity, and these are transmitted to the microphone diaphragm where they are converted to an electrical signal. A variation of the air-coupled sensor arrangement is a commercial stethoscope chest piece connected by a short piece of plastic tubing to a microphone element. Air-coupled microphones have become popular for recording lung sounds because they are simple, inexpensive (most under $50.00), and easily powered, and they adapt to commonly available recording devices. However, they are not intended for such encapsulated use, they are not characterized for such use by the manufacturer, and there is no device (comparable to a shaker table for accelerometers) to evaluate their behavior in actual use.

Commercial electronic stethoscopes have usually been reserved for clinical use and not often employed in lung sound research. Nevertheless, the Andries Tek stethoscope (Andries Tek, Austin, TX) has been used in several studies by one group during the past decade (3, 5, 33). The Andries Tek stethoscope uses a bell-shaped chamber as a microphone coupler and has no diaphragm or adjustable audio filter.

The object of the present study was to compare the frequency responses of a sample of sensors frequently used during the past decade in lung sound research. We did not attempt to duplicate lung sounds for this comparison but rather used a system described below as a stable generator of sound signals with frequency components encompassing all those present in normal and adventitious lung sounds, especially between 200 and 1,200 Hz where lung sounds contain most recognizable energy.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
We used a test system called the Bioacoustic Transducer Tester (BATT), which has been previously described (18). Briefly, it is composed of a speaker in a rigid enclosure covered at the top by a viscoelastic polyurethane polymer surface made of Akton (Action Products, Hagerstown, MD). This surface emulates the mechanical properties of skin and subcutaneous tissue. In a previous test, we compared Akton to similar thicknesses of fresh meat and fat and found the acoustic properties to be similar (18).

For the testing described here, we compared a variety of transducers that have commonly been used in lung sound research over the past 10 yr (Tables 1 and 2, Fig. 1). We used both white noise and short pulses to emulate continuous and discontinuous lung sounds, respectively. Both types of sound were produced by a synthesized signal generator (model SRC20, Larson Davis, Provo, UT). Comparisons were made with published studies using human lung sounds when such were available (Table 1, protocol A). All of the sensors and couplers were owned by the investigators and were used in the manner commonly employed for lung sound studies. The one exception was the PPG 201 that was used without the associated PPG amplifier and band-pass filters. We wanted to evaluate the characteristics of all the devices in the absence of external filtration.


View this table:
[in this window]
[in a new window]
  		Table 1. Protocols for white noise testing

 

View this table:
[in this window]
[in a new window]
  		Table 2. Sensors used for recording pulses

 

Figure 1
View larger version (40K):
[in this window]
[in a new window]
  		Fig. 1. Photographs of different sensor packages displayed at same scale. Each of these sensors has been used in published lung sound studies over the past 10 yr.

 
To determine the frequency response, pseudorandom white noise (flat within ±2 dB between 30 and 4,000 Hz) was amplified and supplied to the speaker of the BATT. The sound amplitude was adjusted to yield a signal-to-noise ratio of ~60 dB for each sensor at the surface of the Akton layer. At this input level, the sound was just loud enough to be heard by ear at a distance of a few inches from the BATT in a quiet room, and the sensor outputs were judged to be ~6 dB higher than tracheal sounds recorded from the suprasternal notch during deep but not forced breathing. Sound was recorded from the surface of the BATT by whichever device was being tested and also from the chamber beneath the Akton surface. The within-chamber sound was recorded by an electret condenser microphone (model 33-1052, Radio Shack) sealed in a 3-ml syringe tip inserted into a hole in the side of the chamber. This was done both to monitor the sound signature within the chamber as an index of stability of the interior dimensions of the BATT throughout the course of the study and to compare the internal sound signature with that of the surface sound for selected parts of the study.

The pulses used as crackle surrogates consisted of a train of single, 1,000-Hz waveforms generated at a frequency of 10/s. The pulse shape, originally a single, 1,000-Hz wave, became more complex because of the unavoidable resonances within the BATT cavity. We considered this complex subsurface waveform to be the "real" shape of the pulse against which the pulse shape at the surface would be compared. The ability of the surface sensors to faithfully reproduce the shape of the subsurface pulse waveform was evaluated by cross-correlating a 1-s segment of the pulse train (10 pulses) as acquired at the surface against the corresponding pulse train of the input (chamber). The absolute peak normalized correlation coefficient represented the degree of similarity between the pairs of signals above and below the Akton surface: 1.0 = identical, 0.0 = completely dissimilar. We also measured the initial deflection width of the pulses because this is the most frequently evaluated feature of crackles and is often used to discriminate between fine and coarse crackles (1, 14, 42). The duration of the first two cycles of the pulses, another frequently assessed crackle parameter, was not measured because not all sensors displayed two cycles.

The project was carried out at two different times and locations using similar procedures and equipment as described below. Site 1 was within an anechoic chamber at Purdue University. Site 2 was in a quiet room at the University of Kentucky. In both instances, the BATT device was supported by a foam vibration isolation platform to attenuate floor and table vibrations. The center of the Akton surface was marked with indelible ink, and each sensor was placed precisely at the center of the test surface. All sensors except the Littman stethoscope head and Andries Tek stethoscope (that were held by gravity) were fastened to the surface of the BATT using a double adhesive tape ring because this is the way they are generally used in practice.

Signal Handling

Site 1.   The electrical output of the microphones used for the evaluation was preamplified by a stereo mixer (Realistic model 32-1100A, Tandy, Ft. Worth, TX) before digitization by a data-acquisition card (model PCI-MIO-16E-1, National Instruments). The Siemens accelerometer was used with its associated electronics before its electrical output was digitized and saved. The signals were digitized at 22,500 samples/s with 16-bit amplitude resolution. A fast Fourier transform (FFT) was performed on 1,024-point epochs of data after multiplying by a Hanning window with 50% overlap between epochs. Fifty records were averaged in the frequency domain to yield the final spectrum.

Site 2.   After preamplification by the same stereo mixer as at site 1, the sounds were digitized at 44,100 Hz by an iMac computer with 16-bit amplitude resolution. A fast Fourier transform was performed on 1,024-point epochs of data after multiplying by a Hanning window, with 0% overlap between epochs. Fifty records were averaged in the frequency domain to yield the final spectrum.

The resulting averaged spectra were converted to decibels with an arbitrary reference value. These averaged spectra were saved and imported into a statistical and graphing program (JMP version 5, SAS Institute, Cary, NC) for display.

We compared the spectra from the different sensors or same sensors with different couplers as outlined in Table 1. The pulses were compared as outlined in Table 2.


   RESULTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 GRANTS
 REFERENCES
 
Figure 2 shows the comparative spectral shapes of white noise as detected by the five different sensors applied to the surface of the BATT. The frequency range of most clinically evident lung sounds is between 200 and 1,000 Hz. All of the sensors were sensitive in this frequency range, although the smoothness of their frequency responses varied considerably. Certain respiratory sounds, most notably tracheal sounds and fine crackles, have components above 1,000 Hz. All of the sensors have some sensitivity up to 2,000 Hz except for the Andries Tek. The PPG is notable for displaying relatively high sensitivity up to 4,000 Hz.


Figure 2
View larger version (36K):
[in this window]
[in a new window]
  		Fig. 2. Comparison of the frequency dependent sensitivity of the 5 types of lung sound transducers displayed on the same amplitude and frequency scale. The spectra are intentionally offset on the y-axis to allow them to be individually distinguishable.

 
Figure 3 shows comparisons of the air chamber coupler under different conditions. Figure 3A shows the effects of different widths of the chamber coupler. These couplers were the same as were used in our previous laboratory's study in which human lung sounds were the test signal (19). We believe that the differences seen here were insignificant (<3 dB). Trivially small differences were also seen between conical and cylindrical chamber shapes (Fig. 3B). Figure 3C shows the effect of a 0.6-mm-diameter vent on the frequency response. As expected, this affected the lower frequencies much more than higher frequencies because the viscosity of the air in the vent resists high-frequency fluctuations more than low.


Figure 3
View larger version (24K):
[in this window]
[in a new window]
  		Fig. 3. A: spectra of white noise signal recorded from the surface of the BATT by a Sony ECM-T150 electret condenser microphone inserted in plastic couplers with cylindrical air chambers of 2-mm depth and 3 different diameters (5, 10, and 15 mm). The relative amplitudes displayed on the y-axis are as actually recorded (no offset applied). B: spectra of white noise signal recorded from surface of BATT by a Sony ECM-T150 electret condenser microphone inserted in 2 couplers with air chambers of 10-mm diameter and 2-mm depth, one cylindrically shaped (Cyl) and the other conically shaped (Con). C: spectra of white noise signal recorded frm surface of BATT by Sony ECM-T150 electret condenser microphone inserted in plastic coupler with and without a 0.6-mm-diameter vent.

 
Figure 4 shows the surface spectrum acquired by the Siemens and PPG accelerometers compared with the noise spectrum within the chamber beneath the surface. The chamber spectra are similar in both cases. The PPG accelerometer was superior in accurately detecting the higher frequencies of the chamber sound from its location on the Akton surface.


Figure 4
View larger version (32K):
[in this window]
[in a new window]
  		Fig. 4. White noise spectra from surface and chamber of BATT recorded by Siemens (top) and PPG (bottom) accelerometers.

 
Figure 5 shows the performance of the Littman stethoscope head in its diaphragm and bell configurations. The bell's relative attenuation of frequencies over 400 Hz compared with the diaphragm is evident.


Figure 5
View larger version (43K):
[in this window]
[in a new window]
  		Fig. 5. Spectra of white noise signal recorded from the surface of the BATT by diaphragm and bell of Littman stethoscope head connected to a Radio Shack electret condenser microphone by a 2-cm length of plastic tubing. The relative amplitudes as displayed on the y-axis are as actually recorded (no offset applied).

 
Figure 6 shows the response of the sensors to pulses. As in the comparisons of white noise spectra, the PPG performed the best in accurately reproducing the shape of the pulse that was present below the surface layer, although the Siemens and Sony microphone in the air chamber coupler also performed well.


Figure 6
View larger version (14K):
[in this window]
[in a new window]
  		Fig. 6. Accuracy with which each type of lung sound transduction device captures the waveform of a 1,000-Hz pulse fed to the speaker of the BATT. The waveform designated "chamber" is the shape of the pulse within the BATT underneath the Akton surface. IDW, initial deflection width. The normalized cross-correlation coefficient compares the phase and timing of the waveform deflections at the surface with those of the waveform within the chamber. A cross-correlation coefficient of 1.0 indicates identical signal shape, whereas 0.0 indicates complete dissimilarity.

 

   DISCUSSION
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
GRANTS
 REFERENCES
 
The theory underlying lung sound transduction and the relative advantages of different transducers is beyond the scope of this article and has been discussed at length by others (9). It is not always necessary to accurately capture all the spectral components of lung sound signals. Often, only amplitude variations are of interest, and, in those cases, it is sufficient to have sensitivity to the louder sounds produced by breathing such as tracheal sounds as surrogates for airflow or tracheal configuration (10, 11, 40, 32), or to map amplitude variations (29, 30). Similarly, the detection of wheezing does not require a wide bandwidth, and the Andries Tek stethoscope, despite its limited frequency response, has been used successfully for this purpose (4, 5). The same applies to detection of crackles (27, 39).

Other types of studies require broad acoustic sensitivity such as analyses of crackle waveforms (2, 25, 26, 41), the measurement of sound transmitted through airways (6, 15, 20, 38, 44, 46, 47), and measurement of tracheal spectral irregularities, especially at frequencies above 1,000 Hz (16, 17, 35–37, 48).

Reproducibility is the main advantage of evaluating lung sound sensors with an artificial chest wall and stable sound source. This is not possible when using humans as lung sound sources because of the innate variability of breathing and the presence of other biological sounds such as muscle, cardiovascular, and intestinal noises. Although sensors can be placed side by side on the chest wall, the well-established positional heterogeneity (7) of lung sounds adds poorly defined complexity to the analysis. The BATT makes these sensor comparisons relatively easy by removing biological variability while allowing one to test both microphone-based sensors and accelerometers. The test sounds are intrinsically stable, are reproducible, and encompass all the frequencies of interest to lung sound researchers. The comparisons reported here reveal large differences in the shape of the spectra captured by the different sensors. One unavoidable conclusion is that these sensors are, in general, not interchangeable, and we would expect that sounds accessed by different sensors would appear different.

The PPG 201 deserves special mention. It is an accelerometer that has been designed and used exclusively for lung sound research. It contains no internal filter and relies on external filtration to counteract its characteristically uneven frequency response. According to the PPG's specifications, the piezoelectric element resonates at 3,700 Hz, and at that frequency the device is 100 times more sensitive than it is at 1,000 Hz. This idiosyncrasy makes the PPG 201 more sensitive to frequencies above 1,200 Hz than any of the other sensors evaluated, and this is probably responsible for the ability of the PPG to accurately reproduce pulse waveforms. The amplifier-filter that is intended to be used with the PPG blunts this uneven frequency response, making it flatter. Nevertheless, the natural high-frequency emphasis of the unfiltered PPG seems well suited for sounds with high-frequency components such as crackles. However, its low-frequency performance appears to suffer compared with all the other tested sensors as shown in Fig. 2 with an approximately linear drop of 12 dB from 1,000 Hz down to 200 Hz.

Examination of Fig. 2 reveals that, for frequencies between 200 and 1,200 Hz, the Siemens accelerometer, air chamber coupler, and Littman diaphragm arrangement provide reasonably consistent frequency responses comparable within ±5 dB. We expect that these three devices would perform similarly with normal lung sounds. The Andries Tek electronic stethoscope is quite sensitive to the loudest components of the vesicular sound at ~200 Hz but then peaks in sensitivity at 400 Hz and rapidly falls off above that. It would suffer when presented with the higher frequency components of tracheal sounds that extend to above 1,500 Hz (17). As a general purpose lung sound sensor, we find the Andries Tek to be least useful.

How much do the differences between these sensors matter when presented with real respiratory sounds rather than white noise? The usefulness of any tool depends on how appropriately it is matched to the job it is expected to do. The Andries Tek electronic stethoscope has a much more constrained frequency response than any of the other sensors, but it appears reasonably sensitive to sound between 200 and ~800 Hz that encompasses most of the frequency components of the vesicular sound. To illustrate this, we compared the Sony ECM-T150 microphone in the air-coupled chest piece and the Andries Tek electronic stethoscope with vesicular and tracheal sounds instead of white noise played through the BATT. These sounds of normal breathing were made by recording lung sounds from the left anterior chest and from over the trachea at the suprasternal notch of one of the authors (S. S. Kraman). The waveforms were displayed on a computer screen, and a section, ~0.3 s long, of the highest amplitude portion of each of the inspiratory sounds were selected and copied to the computer buffer. The sound segments were copied and concatenated (perfectly matching the beginning and end of each copied segment) to produce two 10-s sound files containing, respectively, the spectral components of a vesicular sound and a tracheal sound without the cyclic, flow-dependent amplitude variations inherent in breathing sounds. These sounds were then played into the BATT and recorded from the surface by each of the sensors. Averaged sound spectra were then created (Fig. 7). Despite some differences, the vesicular sound spectrum appears similar with both sensors. The Andries Tek, however, failed to adequately detect the higher frequency components of the tracheal sound such as the resonances at 1,200 and 1,600 Hz when compared with the air-coupled sensor.


Figure 7
View larger version (21K):
[in this window]
[in a new window]
  		Fig. 7. Vesicular and tracheal sounds recorded from the surface of the BATT using an air-coupled Sony ECM-150 microphone (air coupled) and the Andries Tek (Andries) electronic stethoscope. Note the decreased sensitivity of the Andries Tek stethoscope above 400 Hz and virtual insensitivity above 1,000 Hz.

 
We conclude that, although all the evaluated sensors were adequate for at least some types of lung sound research, they differ significantly in their ability to capture the spectral components and waveforms of these sounds. Characterization of the idiosyncrasies of these sensors begins to make comparisons of lung sounds from among different laboratories feasible.


   GRANTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
REFERENCES
 
S. S. Kraman is supported by the Margaret Logan Colvin Chair in Lung Disease Research


   FOOTNOTES
 

Address for reprint requests and other correspondence: S. S. Kraman, Univ. of Kentucky, Kentucky Clinic L-547, Lexington, KY 40536 (e-mail: sskram01{at}uky.edu)

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


   REFERENCES
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 

  1. Al Jarad N, Davies SW, Logan-Sinclair R, and Rudd RM. Lung crackle characteristics in patients with asbestosis, asbestos-related pleural disease and left ventricular failure using a time-expanded waveform analysis—a comparative study. Respir Med 88: 37–46, 1994.[CrossRef][Web of Science][Medline]
  2. Al Jarad N, Strickland B, Bothamley G, Lock S, Logan-Sinclair R, and Rudd RM. Diagnosis of asbestosis by a time expanded wave form analysis, auscultation and high resolution computed tomography: a comparative study. Thorax 48: 347–353, 1993.[Abstract/Free Full Text]
  3. Bohadana AB, Kopferschmitt-Kubler MC, and Pauli G. Breath sound intensity in patients with airway provocation challenge test positive by spirometry but negative for wheezing: a preliminary report. Respiration 61: 274–279, 1994.[Web of Science][Medline]
  4. Bohadana AB, Massin N, Teculescu D, and Peslin R. Tracheal wheezes during methacholine airway challenge (MAC) in workers exposed to occupational hazards. Respir Med 88: 581–587, 1994.[Web of Science][Medline]
  5. Bohadana AB, Peslin R, Uffholtz H, and Pauli G. Potential for lung sound monitoring during bronchial provocation testing. Thorax 50: 955–961, 1995.[Abstract/Free Full Text]
  6. Donnerberg RL, Druzgalski CK, Hamlin RL, Davis GL, Campbell RM, and Rice DA. Sound transfer function of the congested canine lung. Br J Dis Chest 74: 23–31, 1980.[Web of Science][Medline]
  7. Dosani R and Kraman SS. Lung sound intensity variability in normal men. A contour phonopneumographic study. Chest 83: 628–631, 1983.[Abstract/Free Full Text]
  8. Forgacs P. Lung sounds. Br J Dis Chest 63: 1–12, 1969.[CrossRef][Medline]
  9. Gavriely N. Breath Sounds Methodology. Boca Raton, FL: CRC, 1995.
  10. Harper VP, Pasterkamp H, Kiyokawa H, and Wodicka GR. Modeling and measurement of flow effects on tracheal sounds. IEEE Trans Biomed Eng 50: 1–10, 2003.[CrossRef][Web of Science][Medline]
  11. Hida W, Miki H, Kikuchi Y, Miura C, Iwase N, Shimizu Y, and Takishima T. Home sleep monitor for detecting apnea episodes by nasal flow and tracheal sound recordings. Tohoku J Exp Med 156, Suppl: 137–142, 1988.
  12. Jalil A and Aronovitch M. Recording and analysis of breath sounds in normal individuals. Indian J Chest Dis 8: 158–166 concl, 1966.[Medline]
  13. Jalil A and Aronovitch M. Recording and analysis of breath sounds in patients of chest diseases. II. Indian J Chest Dis 8: 188–194, 1966.[Medline]
  14. Kawamura T, Matsumoto T, Tanaka N, Kido S, Jiang Z, and Matsunaga N. Crackle analysis for chest auscultation and comparison with high-resolution CT findings. Radiat Med 21: 258–266, 2003.[Medline]
  15. Kraman SS and Bohadana AB. Transmission to the chest of sound introduced at the mouth. J Appl Physiol 66: 278–281, 1989.[Abstract/Free Full Text]
  16. Kraman SS, Harper P, Pasterkamp H, and Wodicka GR. ‘Slide whistle' breath sounds: acoustical correlates of variable tracheal obstruction. Physiol Meas 23: 449–455, 2002.[CrossRef][Web of Science][Medline]
  17. Kraman SS, Pasterkamp H, Kompis M, Takase M, and Wodicka GR. Effects of breathing pathways on tracheal sound spectral features. Respir Physiol 111: 295–300, 1998.[CrossRef][Web of Science][Medline]
  18. Kraman SS, Pressler GA, Pasterkamp H, and Wodicka GR. Design, construction, and evaluation of a BioAcoustic Transducer Testing (BATT) system for respiratory sounds. IEEE Trans Biomed Eng. In press.
  19. Kraman SS, Wodicka GR, Oh Y, and Pasterkamp H. Measurement of respiratory acoustic signals. Effect of microphone air cavity width, shape, and venting. Chest 108: 1004–1008, 1995.[Abstract/Free Full Text]
  20. Kudoh S. [Wave form of intrabronchial spark sound on the chest wall and sound transmission in the lung-thoracic system]. Nippon Ika Daigaku Zasshi-J Nippon Med School 59: 323–334, 1992. [In Japanese]
  21. Loudon R and Murphy RL Jr. Lung sounds (Review). Am Rev Respir Dis 130: 663–673, 1984.[Web of Science][Medline]
  22. McKusick V, Jenkins JT, and Webb GN. The acoustic basis of the chest examination; studies by means of sound spectrography. Am Rev Tuberc 72: 12–34, 1955.[Web of Science][Medline]
  23. McKusick V, Webb GN, Brayshaw JR, and Talbot SA. Spectral phonocardiography: clinical studies. Bull Johns Hopkins Hosp 95: 90–110, 1954.[Medline]
  24. McKusick VA, Massengale ON Jr, Wigod M, and Webb GN. Spectral phonocardiographic studies in congenital heart disease. Br Heart J 18: 403–416, 1956.
  25. Mori M, Kinoshita K, Morinari H, Shiraishi T, Koike S, and Murao S. Waveform and spectral analysis of crackles. Thorax 35: 843–850, 1980.[Abstract/Free Full Text]
  26. Munakata M, Ukita H, Doi I, Ohtsuka Y, Masaki Y, Homma Y, and Kawakami Y. Spectral and waveform characteristics of fine and coarse crackles. Thorax 46: 651–657, 1991.[Abstract/Free Full Text]
  27. Murphy RL Jr, Gaensler EA, Holford SK, Del Bono EA, and Epler G. Crackles in the early detection of asbestosis. Am Rev Respir Dis 129: 375–379, 1984.[Web of Science][Medline]
  28. Nairn JR, and Turner-Warwick M. Breath sounds in emphysema. Br J Dis Chest 63: 29–37, 1969.[CrossRef][Medline]
  29. O'Donnell DM and Kraman SS. Vesicular lung sound amplitude mapping by automated flow-gated phonopneumography. J Appl Physiol 53: 603–609, 1982.[Abstract/Free Full Text]
  30. Pasterkamp H, Consunji-Araneta R, Oh Y, and Holbrow J. Chest surface mapping of lung sounds during methacholine challenge. Pediatr Pulmonol 23: 21–30, 1997.[CrossRef][Web of Science][Medline]
  31. Pasterkamp H, Kraman SS, DeFrain PD, and Wodicka GR. Measurement of respiratory acoustical signals. Comparison of sensors. Chest 104: 1518–1525, 1993.[Abstract/Free Full Text]
  32. Pasterkamp H, Schafer J, and Wodicka GR. Posture-dependent change of tracheal sounds at standardized flows in patients with obstructive sleep apnea. Chest 110: 1493–1498, 1996.[Abstract/Free Full Text]
  33. Purohit A, Bohadana A, Kopferschmitt-Kubler MC, Mahr L, Linder J, and Pauli G. Lung auscultation in airway challenge testing. Respir Med 91: 151–157, 1997.[CrossRef][Web of Science][Medline]
  34. Rozenblat VV, Likhacheva EI, Stolbun BM, Sazhina TG, Forshtadt VM, and Ganiushkina SM. [The frequency spectrum of respiratory sounds in normal subjects and in pneumoconiosis]. Klin Med (Mosk) 46: 135–139, 1968.
  35. Sanchez I and Pasterkamp H. Tracheal sound spectra depend on body height. Am Rev Respir Dis 148: 1083–1087, 1993.[Web of Science][Medline]
  36. Sanchez I and Vizcaya C. Tracheal and lung sounds repeatability in normal adults. Respir Med 97: 1257–1260, 2003.[CrossRef][Web of Science][Medline]
  37. Sano K, Nakano H, Ohnishi Y, Ishii Y, Maekawa J, Nakatani Y, and Narita N. [Changes in tracheal and lung sounds before and after resection of adenoid cystic carcinoma of the trachea]. Nihon Kokyuki Gakkai Zasshi 37: 987–991, 1999.[Medline]
  38. Shioya N, Takezawa Y, Mikami R, Kudoh S, and Shibuya A. [Acoustic transmission of the respiratory system using orally introduced sinusoidal sound waves]. Nippon Kyobu Shikkan Gakkai Zasshi 22: 125–130, 1984.
  39. Shirai F, Kudoh S, Shibuya A, Sada K, and Mikami R. Crackles in asbestos workers: auscultation and lung sound analysis. Br J Dis Chest 75: 386–396, 1981.[CrossRef][Web of Science][Medline]
  40. Soufflet G, Charbonneau G, Polit M, Attal P, Denjean A, Escourrou P, and Gaultier C. Interaction between tracheal sound and flow rate: a comparison of some different flow evaluations from lung sounds. IEEE Trans Biomed Eng 37: 384–391, 1990.[CrossRef][Web of Science][Medline]
  41. Sovijarvi AR, Helisto P, Malmberg LP, Kallio K, Paajanen E, Saarinen A, Lipponen P, Haltsonen S, Pekkanen L, Piirila P, Naveri L, and Katila T. A new versatile PC-based lung sound analyzer with automatic crackle analysis (HeLSA); repeatability of spectral parameters and sound amplitude in healthy subjects. Technol Health Care 6: 11–22, 1998.[Medline]
  42. Sovijarvi AR, Piirila P, and Luukkonen R. Separation of pulmonary disorders with two-dimensional discriminant analysis of crackles. Clin Physiol 16: 171–181, 1996.[Web of Science][Medline]
  43. Weiss EB and Carlson CJ. Recording of breath sounds. Am Rev Respir Dis 105: 835–839, 1972.[Web of Science][Medline]
  44. Wodicka GR, DeFrain PD, and Kraman SS. Bilateral asymmetry of respiratory acoustic transmission. Med Biol Eng Comput 32: 489–494, 1994.[Web of Science][Medline]
  45. Wodicka GR, Kraman SS, Zenk GM, and Pasterkamp H. Measurement of respiratory acoustic signals. Effect of microphone air cavity depth. Chest 106: 1140–1144, 1994.[Abstract/Free Full Text]
  46. Wodicka GR and Shannon DC. Transfer function of sound transmission in subglottal human respiratory system at low frequencies. J Appl Physiol 69: 2126–2130, 1990.[Abstract/Free Full Text]
  47. Wodicka GR, Stevens KN, Golub HL, and Shannon DC. Spectral characteristics of sound transmission in the human respiratory system. IEEE Trans Biomed Eng 37: 1130–1135, 1990.[CrossRef][Web of Science][Medline]
  48. Yonemaru M, Kikuchi K, Mori M, Kawai A, Abe T, Kawashiro T, Ishihara T, and Yokoyama T. Detection of tracheal stenosis by frequency analysis of tracheal sounds. J Appl Physiol 75: 605–612, 1993.[Abstract/Free Full Text]
  49. Zislin DM, Rozenblat VV, and Likhacheva EI. [Objective study of respiratory sounds using frequency analysis]. Ter Arkh 41: 108–113, 1969.[Medline]



This article has been cited by other articles:


Home page
 Postgrad. Med. J.Home page
C. C Ceresa and I. D A Johnston
Auscultation in the diagnosis of respiratory disease in the 21st century
Postgrad. Med. J., August 1, 2008; 84(994): 393 - 394.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow All Versions of this Article:
101/2/469    most recent
00273.2006v1
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Web of Science (4)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kraman, S. S.
Right arrow Articles by Pasterkamp, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kraman, S. S.
Right arrow Articles by Pasterkamp, H.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online
Copyright © 2006 by the American Physiological Society.