Ipek Sen

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

Classification of respiratory signals by linear analysis

The aim of this study is the classification of wheeze and non-wheeze epochs within respiratory sound signals acquired from patients with asthma and COPD. Since a wheeze signal, having a sinusoidal waveform, has a different behavior in time and frequency domains from that of a non-wheeze signal, the features selected for classification are kurtosis, Renyi entropy, f50/ f90 ratio and mean-crossing irregularity. Upon calculation of these features for each wheeze and non-wheeze portion, the whole data scattered as two classes in four dimensional feature space is projected using Fisher Discriminant Analysis (FDA) onto the single dimensional space that separates the two classes best. Observing that the two classes are visually well separated in this new space, Neyman-Pearson hypothesis testing is applied. Finally, the correct classification rate is %95.1 for the training set, and leave-one-out approach pursuing the above methodology yields a success rate of %93.5 for the test set.

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.

A Comparison of SVM and GMM-Based Classifier Configurations for Diagnostic Classification of Pulmonary Sounds

Goal: The aim of this study is to find a useful methodology to classify multiple distinct pulmonary conditions including the healthy condition and various pathological types, using pulmonary sounds data. Methods: Fourteen-channel pulmonary sounds data of 40 subjects (healthy and pathological, where the pathologies are of obstructive and restrictive types) are modeled using a second order 250-point vector autoregressive model. The estimated model parameters are fed to support vector machine and Gaussian mixture model (GMM) classifiers which are used in various configurations, resulting in eight different methodologies in total. Results: Among the eight methodologies, the hierarchical GMM classifier yields the best performance with a total correct classification rate of 85%, where the term hierarchical refers here to first classifying the data into healthy and pathological classes, then the pathological class into obstructive and restrictive types. Both the sensitivity and specificity for the healthy versus pathological classification at the first stage of hierarchy are 90%. Conclusion: The newly proposed methodologies provide improved results compared to the previous study. The hierarchical framework is suggested for diagnostic classification of pulmonary sounds, although the algorithms are still open for further improvements. Significance: This study proposes new methodologies for diagnostic classification of pulmonary sounds, and suggests using a hierarchical framework for the first time.

Exploring an optimal vector autoregressive model for multi-channel pulmonary sound data

The purpose of this study is to find a useful mathematical model for multi-channel pulmonary sound data. Vector auto-regressive (VAR) model schema is adopted and the best set of arguments, namely, the order and sample size of the model and the sampling rate of the data, is aimed to be determined. Both conventional prediction error criteria and a set of three new criteria which are derived specifically for pulmonary sound signals are used to evaluate the success of the model. In terms of these criteria, the second order 250-point model is selected to be the most descriptive VAR model for 14-channel pulmonary sound data. The preferred sampling rate is the original data acquisition rate, which is 9600 samples per second. The effect of normalization of the data with respect to the air flow is also examined. Six normalization schemes are implemented on the data prior to the model fit, and the resulting model parameters are examined in terms of the proposed criterion measures. It is concluded that normalization with respect to flow is not necessary prior to the VAR modeling of pulmonary sound data.

Acoustic mapping of the lung based on source localization of adventitious respiratory sound components

The aim of this study is to devise a methodology to estimate and depict the source locations of respiratory adventitious sound components in the lungs, particularly crackles, associated with certain pulmonary diseases. Using the multichannel respiratory sound signals recorded on the chest wall, we have tried to locate the sources of crackling sounds. The source localization is performed using basic independent component analysis (basic ICA) followed by an evaluation of the mixing coefficients in a center of weights approach, where after the ICA, by taking the relevant mixing matrix coefficients and assuming them to be placed on the microphone locations, the estimated sound source location is calculated as the center of those weights. In order to select both the proper data segments prior to the ICA, and the relevant independent component (IC) among the source signal estimates of the ICA subsequently, a Bayesian classifier (under the assumption of Gaussian likelihoods) has been trained, using the data of the same subject yet a different acquisition session from the one intended for source localization. The outcome of the algorithm is a map of estimated source locations of crackles with respect to the microphone locations, which is presented together with the error performances (both validation and test) of the classifier. This approach for the estimation and mapping of the adventitious sound source locations in the lungs using the acoustic data may be a promising imaging alternative, which is practical, non-expensive and harmless.

A lung sound classification system based on the rational dilation wavelet transform

In this work, a wavelet based classification system that aims to discriminate crackle, normal and wheeze lung sounds is presented. While the previous works related with this problem use constant low Q-factor wavelets, which have limited frequency resolution and can not cope with oscillatory signals, in the proposed system, the Rational Dilation Wavelet Transform, whose Q-factors can be tuned, is employed. Proposed system yields an accuracy of 95 % for crackle, 97 % for wheeze, 93.50 % for normal and 95.17 % for total sound signal types using energy feature subset and proposed approach is superior to conventional low Q-factor wavelet analysis.

Feature extraction using time-frequency analysis for monophonic-polyphonic wheeze discrimination

The aim of this study is monophonic-polyphonic wheeze episode discrimination rather than the conventional wheeze (versus non-wheeze) episode detection. We used two different methods for feature extraction to discriminate monophonic and polyphonic wheeze episodes. One of the methods is based on frequency analysis and the other is based on time analysis. Frequency analysis based method uses ratios of quartile frequencies to exploit the difference in the power spectrum. Time analysis based method uses mean crossing irregularity to exploit the difference in periodicity in the time domain. Both methods are applied on the data before and after an image processing based preprocessing step. Calculated features are used in classification both individually and in combinations. Support vector machine, k-nearest neighbor and Naive Bayesian classifiers are adopted in leave-one-out scheme. A total of 121 monophonic and 110 polyphonic wheeze episodes are used in the experiments, where the best classification performances are 71.45% for time domain based features, 68.43% for frequency domain based features, and 75.78% for a combination of selected best features.

Adaptive Modeling of Sound Transmission in the Respiratory System

In this study, adaptive filtering techniques have been used in an attempt to model the respiratory system. The respiratory system has been considered as a dynamic system for which input-output relationship is to be defined. Simultaneous measurement of the respiratory sounds over the trachea and posterior chest were made, with the signal from the trachea forming the input to a finite impulse response filter and the signal from the posterior chest forming the desired response of the filter. The chest cavity was stimulated with speech sounds. Least-mean square algorithm was used to update filter coefficients. The learning curves of the filter are presented in the paper. It can be concluded that adaptive filtering is a promising way to characterize transmission characteristics of the respiratory system and further improvement may be obtained if anatomical information is integrated in the modeling process.

A multichannel analog processing circuit for respiratory sound acquisition applications

A five-channel analog amplifier & filter stage of pulmonary sound data acquisition process has been designed and implemented. The system acquires respiratory sound data through five microphones placed at proper locations on the chest wall along with a flow signal for synchronization. Output is band limited and amplified sound data suitable for further analog-to-digital (AD) conversion. The gain and the high corner frequency are adjustable, the former aiming an optimum use of the AD converter by approximating maximum possible input amplitude and the latter aiming a provision for a possible change to be made in the sampling frequency of the AD converter. In the design, low power consumption, low noise, low cost and low component count have been taken into consideration.