Validation of manual and automated wheezing detection from audio recordings
Bibliographic record
Abstract
Introduction: Presently human experts must diagnose wheezing at the bedside using real-time analog tools (ear, stethoscope). Human experts could diagnose wheezing from digital audio files. Further, automated wheezing detection from unattended digital recordings would obviate bedside human diagnosis. Aim: To validate the detection of wheezing from unattended digital recordings. Methods: 189 digital audio recordings of 30 s duration were obtained using a dedicated digital microphone device (Wheezo, Respiri Limited, Melbourne) from 56 hospitalized patients (38 F; age 21 to 87 - mean 63.5 years; 26 COPD, 27 asthma, 1 vocal cord dysfunction, and 2 other) and 20 ambulatory normal adult controls. Wheezes were scored manually by two respirologists (KP, NS) and by a biomedical engineer (RSP). Their consensus was compared to an automated wheeze scoring algorithm. The accuracy, sensitivity, and specificity of automated wheeze rate detection were calculated along with Cohen9s κ coefficient (R 3.4.4). We further plan to apply the algorithm to the ERS reference database of lung sounds. Results: The accuracy, sensitivity, and specificity of automated wheeze rate detection were 90.5, 87.1, and 93.3%. Cohen9s κ coefficient was 0.81. Breath-by-breath comparison of human and automated scoring showed near-perfect agreement for the presence, absence, and duration of wheezes. Conclusion: Wheezing can be diagnosed from digital audio recordings either manually by human experts or automatically by a computer algorithm. Diagnosing wheezing remotely increases assessment capacity while reducing viral transmission risk during epidemics.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.014 | 0.030 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.000 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".