Technology and health care : official journal of the European Society for Engineering and Medicine
-
Technol Health Care · Jun 1998
Classification of respiratory sounds based on wavelet packet decomposition and learning vector quantization.
In this paper, a wavelet packet-based method is used for detection of abnormal respiratory sounds. The sound signal is divided into segments, and a feature vector for classification is formed using the results of the search for the best wavelet packet decomposition. ⋯ The method is tested using a small set of real patient data which was also analysed by an expert observer. The preliminary results are promising, although not yet good enough for clinical use.
-
Technol Health Care · Jun 1998
A new versatile PC-based lung sound analyzer with automatic crackle analysis (HeLSA); repeatability of spectral parameters and sound amplitude in healthy subjects.
A versatile PC-based lung sound analyzer has been developed for short-term recording and analysis of respiratory sounds in research and clinical applications. The system consists of two sound sensors, a flow sensor, a filtering signal amplifier and a PC with a data acquisition card and software for measurement and analysis of the sounds. The analyses include phonopneumography, time expanded waveform analysis, spectral analysis with time averaged Fast Fourier Transform, frequency analysis in time domain (sonogram), and automatic detection and waveform analysis of crackles. ⋯ Examples of lung sound analysis of samples containing adventitious sounds such as crackles and wheezes are presented. The results indicate that the median frequency has the best repeatability of quartile frequencies of breath sounds and they suggest that the variations of those parameters are low enough for diagnostic purposes. The results also suggest that the analyzer can be a useful new tool for pulmonary research in the fields of physiological and clinical short-term studies of respiratory sounds.
-
A new automatic wheeze detection method which is based on image processing techniques applied to the sonagram was developed here. In the calculation of the sonagram, autoregressive and FFT spectrum estimation methods were compared. The method was validated in four wheezing asthmatic patients by a pulmonary physician. ⋯ Very short wheezes were not detected. The false positive amount of wheezing in control subjects was only about 1%. The method extracts also information about the frequency, duration, flow and volume associated with the wheezes.