A multiresolution analysis for detection of abnormal lung sounds

Abstract

Automated analysis and detection of abnormal lung sound patterns has great potential for improving access to standardized diagnosis of pulmonary diseases, especially in low-resource settings. In the current study, we develop signal processing tools for analysis of paediatric auscultations recorded under non-ideal noisy conditions. The proposed model is based on a biomimetic multi-resolution analysis of the spectro-temporal modulation details in lung sounds. The methodology provides a detailed description of joint spectral and temporal variations in the signal and proves to be more robust than frequency-based techniques in distinguishing crackles and wheezes from normal breathing sounds.

Fig. 1

Time-frequency (column 1) and rate-scale representation (column 2) of data used in the study, showing a normal(a), a crackle(b) and a wheeze(c) case. Notice the difficulties in discrimination based on only time-freequency contents. However distinction is clearer in the rate-scale representation due to the asymmetry noticed comparing the normal with the adventitious cases and a detailed look would indicate stronger components for crackles.