International perception of lung sounds: a comparison of classification across some European borders

Aviles-Solis, Juan Carlos; Vanbelle, Sophie; Halvorsen, Peder Andreas; Francis, Nick; Cals, Jochen; Andreeva, Elena; Marques, Alda; Piirilä, Päivi; Pasterkamp, Hans; Melbye, Hasse

doi:10.1136/bmjresp-2017-000250

Cited by 26 publications

(34 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As there is no suitable 'gold standard' benchmark for identifying the presence or absence abnormal breath sounds, we used careful audio and spectrogram analysis of real-world, in-hospital recordings to determine this, with validation by a second blinded clinician who applied the same strict labelling criteria demonstrating excellent interrater agreement. Although human expert labelling may introduce potential bias, previous multicenter research has shown that this approach is a method suitable for research of lung sounds [5]. Nevertheless, due to the absence of a true gold standard, we report our results using PPA and NPA, rather than the more familiar terms sensitivity and specificity.…”

Section: Discussionmentioning

confidence: 99%

“…In recent years, stethoscopes capable of digitally recording breath sounds have become more widely available, offering the ability to capture breath sounds with superior sound quality and fidelity [4]. However, human interpretation of the digital recordings can still exhibit significant inter-listener variability [5]. As the soundwave properties of pathologic breath sounds such as crackles, wheezes and rhonchi have been well-studied and previously defined, computer algorithms and programs to automatically detect them have been developed [6,7].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Artificial intelligence accuracy in detecting pathological breath sounds in children using digital stethoscopes

2020

View full text Add to dashboard Cite

Background Manual auscultation to detect abnormal breath sounds has poor inter-observer reliability. Digital stethoscopes with artificial intelligence (AI) could improve reliable detection of these sounds. We aimed to independently test the abilities of AI developed for this purpose. Methods One hundred and ninety two auscultation recordings collected from children using two different digital stethoscopes (Clinicloud™ and Littman™) were each tagged as containing wheezes, crackles or neither by a pediatric respiratory physician, based on audio playback and careful spectrogram and waveform analysis, with a subset validated by a blinded second clinician. These recordings were submitted for analysis by a blinded AI algorithm (StethoMe AI) specifically trained to detect pathologic pediatric breath sounds. Results With optimized AI detection thresholds, crackle detection positive percent agreement (PPA) was 0.95 and negative percent agreement (NPA) was 0.99 for Clinicloud recordings; for Littman-collected sounds PPA was 0.82 and NPA was 0.96. Wheeze detection PPA and NPA were 0.90 and 0.97 respectively (Clinicloud auscultation), with PPA 0.80 and NPA 0.95 for Littman recordings. Conclusions AI can detect crackles and wheeze with a reasonably high degree of accuracy from breath sounds obtained from different digital stethoscope devices, although some device-dependent differences do exist.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Artificial intelligence accuracy in detecting pathological breath sounds in children using digital stethoscopes

2020

View full text Add to dashboard Cite

show abstract

“…For this purpose, an acceptable approach is to ask observers Open access to describe audio or visual displays of lung sounds. [6][7][8][9][10] Using this method, studies of lung sound nomenclature have shown that lung terminology varies widely among physicians, residents and physiotherapists working in the same hospital [6][7][8] or in different countries. 9 Interestingly, agreement on the detailed classification of crackles and wheezes was found to be poor, improving when the two terms were combined into broader categories.…”

Section: Introductionmentioning

confidence: 99%

“…[6][7][8][9][10] Using this method, studies of lung sound nomenclature have shown that lung terminology varies widely among physicians, residents and physiotherapists working in the same hospital [6][7][8] or in different countries. 9 Interestingly, agreement on the detailed classification of crackles and wheezes was found to be poor, improving when the two terms were combined into broader categories. 10 At closer inspection, data from these studies suggest that, while part of the interobserver variation in terminology results from observer preferences (ie, disagreement between correct raters-eg, use of 'rales' or 'crackles'), part might result from lack of auscultatory skill (ie, disagreement due to incorrect sound classification-eg, use of 'normal breath sound' in presence of 'bronchial breathing').…”

Section: Introductionmentioning

confidence: 99%

Influence of observer preferences and auscultatory skill on the choice of terms to describe lung sounds: a survey of staff physicians, residents and medical students

et al. 2020

View full text Add to dashboard Cite

BackgroundIn contrast with the technical progress of the stethoscope, lung sound terminology has remained confused, weakening the usefulness of auscultation. We examined how observer preferences regarding terminology and auscultatory skill influenced the choice of terms used to describe lung sounds.MethodsThirty-one staff physicians (SP), 65 residents (R) and 47 medical students (MS) spontaneously described the audio recordings of 5 lung sounds classified acoustically as: (1) normal breath sound; (2) wheezes; (3) crackles; (4) stridor and (5) pleural friction rub. A rating was considered correct if a correct term or synonym was used to describe it (term use ascribed to preference). The use of any incorrect terms was ascribed to deficient auscultatory skill.ResultsRates of correct sound identification were: (i) normal breath sound: SP=21.4%; R=11.6%; MS=17.1%; (ii) wheezes: SP=82.8%; R=85.2%; MS=86.4%; (iii) crackles: SP=63%; R=68.5%; MS=70.7%; (iv) stridor: SP=92.8%; R=90%; MS=72.1% and (v) pleural friction rub: SP=35.7%; R=6.2%; MS=3.2%. The 3 groups used 66 descriptive terms: 17 were ascribed to preferences regarding terminology, and 49 to deficient auscultatory skill. Three-group agreement on use of a term occurred on 107 occasions: 70 involved correct terms (65.4%) and 37 (34.6%) incorrect ones. Rate of use of recommended terms, rather than accepted synonyms, was 100% for the wheezes and the stridor, 55% for the normal breath sound, 22% for the crackles and 14% for the pleural friction rub.ConclusionsThe observers’ ability to describe lung sounds was high for the wheezes and the stridor, fair for the crackles and poor for the normal breath sound and the pleural friction rub. Lack of auscultatory skill largely surpassed observer preference as a factor determining the choice of terminology. Wide dissemination of educational programs on lung auscultation (eg, self-learning via computer-assisted learning tools) is urgently needed to promote use of standardised lung sound terminology.

show abstract

“…Andrés et al 12 found that the spectrograms Methods Data for classification. The sound recordings employed in this study were also used in the study of Aviles-Solis et al 13 . The description of the sound recording method is therefore the same as in the mentioned study.…”

mentioning

confidence: 99%

The use of spectrograms improves the classification of wheezes and crackles in an educational setting

Aviles-Solis

Storvoll

Vanbelle

et al. 2020

Sci Rep

Self Cite

View full text Add to dashboard Cite

Chest auscultation is a widely used method in the diagnosis of lung diseases. However, the interpretation of lung sounds is a subjective task and disagreements arise. New technological developments like the use of visSual representation of sounds through spectrograms could improve the agreement when classifying lung sounds, but this is not yet known. In this study, we tested if the use of spectrograms improves the agreement when classifying wheezes and crackles. To do this, we asked twenty-three medical students at UiT the Arctic University of Norway to classify 30 lung sounds recordings for the presence of wheezes and crackles. The sample contained 15 normal recordings and 15 with wheezes or crackles. The students classified the recordings in a random order twice. First sound only, then sound with spectrograms. We calculated kappa values for the agreement between each student and the expert classification with and without display of spectrograms and tested for significant improvement between these two coefficients. We also calculated Fleiss kappa for the 23 observers with and without the spectrogram. In an individual analysis comparing each student to an expert annotated reference standard we found that 13 out of 23 students had a positive change in kappa when classifying wheezes with the help of spectrograms. When classifying crackles 16 out of 23 showed improvement when spectrograms were used. In a group analysis we observed that Fleiss kappa values were k = 0.51 and k = 0.56 (p = 0.63) for classifying wheezes without and with spectrograms. For crackles, these values were k = 0.22 and k = 0.40 (p = <0.01) in the same order. Thus, we conclude that the use of spectrograms had a positive impact on the inter-rater agreement and the agreement with experts. We observed a higher improvement in the classification of crackles compared to wheezes.

show abstract

International perception of lung sounds: a comparison of classification across some European borders

Cited by 26 publications

References 34 publications

Artificial intelligence accuracy in detecting pathological breath sounds in children using digital stethoscopes

Artificial intelligence accuracy in detecting pathological breath sounds in children using digital stethoscopes

Influence of observer preferences and auscultatory skill on the choice of terms to describe lung sounds: a survey of staff physicians, residents and medical students

The use of spectrograms improves the classification of wheezes and crackles in an educational setting

Contact Info

Product

Resources

About