Developing a Reference of Normal Lung Sounds in Healthy Peruvian Children
Lung auscultation has long been a standard of care for the diagnosis of respiratory diseases. Recent advances in electronic auscultation and signal processing have yet to find clinical acceptance; however, computerized lung sound analysis may be ideal for pediatric populations in settings, where skilled healthcare providers are commonly unavailable. We described features of normal lung sounds in young children using a novel signal processing approach to lay a foundation for identifying pathologic respiratory sounds.
186 healthy children with normal pulmonary exams and without respiratory complaints were enrolled at a tertiary care hospital in Lima, Peru. Lung sounds were recorded at eight thoracic sites using a digital stethoscope. 151 (81 %) of the recordings were eligible for further analysis. Heavy-crying segments were automatically rejected and features extracted from spectral and temporal signal representations contributed to profiling of lung sounds.
Mean age, height, and weight among study participants were 2.2 years (SD 1.4), 84.7 cm (SD 13.2), and 12.0 kg (SD 3.6), respectively; and, 47 % were boys. We identified ten distinct spectral and spectro-temporal signal parameters and most demonstrated linear relationships with age, height, and weight, while no differences with genders were noted. Older children had a faster decaying spectrum than younger ones. Features like spectral peak width, lower-frequency Mel-frequency cepstral coefficients, and spectro-temporal modulations also showed variations with recording site.
Lung sound extracted features varied significantly with child characteristics and lung site. A comparison with adult studies revealed differences in the extracted features for children. While sound-reduction techniques will improve analysis, we offer a novel, reproducible tool for sound analysis in real-world environments.
KeywordsElectronic auscultation Diagnosis Child Power spectrum Time–frequency analysis Filterbank Spectro-temporal analysis
Additional support came from A.B. PRISMA, Instituto Nacional de Salud del Niño, and collaborators at JHU and Cincinnati Children’s Hospital. Thinklabs Medical (Centennial, CO) generously provided us with electronic stethoscopes at discount. Laura Ellington was supported by the Doris Duke Charitable Foundation Clinical Research Fellowship. Dimitra Emmanouilidou and Mounya Elhilali were partially supported by grants IIS-0846112 (NSF), 1R01AG036424-01 (NIH), N000141010278 (ONR), and N00014-12-1-0740 (ONR). William Checkley and James Tielsch were partially supported by the Bill and Melinda Gates Foundation (OPP1017682).
Conflict of interest
All authors in the study report no conflict of interest.
- 15.Ellington LE, Gilman RH, Tielsch JM, Steinhoff M, Figueroa D, Rodriguez S, Caffo B, Tracey B, Elhilali M, West J, Checkley W (2012) Computerised lung sound analysis to improve the specificity of paediatric pneumonia diagnosis in resource-poor settings: protocol and methods for an observational study. BMJ Open 2:e000506PubMedCentralCrossRefPubMedGoogle Scholar
- 16.Lederman D (2010) Estimation of infants’ cry fundamental frequency using a modified SIFT algorithm. Cornell Univ Online Lib 703–9. http://arxiv.org/abs/1009.2796v1. Accessed 25 Nov 2013
- 19.Boersma P (1993) Accurate short-term analysis of the fundamental frequency and the harmonics-to-noise ratio of a sampled sound. Proc Inst Phon Sci Univ Amst 17:97–110Google Scholar
- 25.Hadjileontiadis LJ (2009) Lung sounds: an advanced signal processing perspective. Morgan & Claypool, San Rafael, CAGoogle Scholar