Simplifying Spirometry: Is one effort enough to rule out obstructive impairment?
Bibliographic record
Abstract
Technically acceptable and repeatable spirometric measures are crucial for accurate pulmonary function test (PFT) interpretation. Artificial intelligence (AI) allows automated quality assessment, indicating reliable measurements in real-time. This study investigated if the first AI-acceptable spirometry trial can rule out obstructive impairment. Spirometry data from the Canadian Longitudinal Study of Aging (Adults aged 45-95) were analysed using ArtiQ.QC software. All trials were assessed using 2019 American Thoracic Society/European Respiratory Society standards. The first acceptable trial (where both forced expiratory volume in one second – FEV1, and Forced Vital Capacity – FVC, were acceptable) and the best acceptable trial (FEV1 and FVC with the highest values) after at least two acceptable and repeatable trials were compared. Race-neutral Global Lung Function Initiative (GLI) lower limits of normal (LLN) classified impairment (5th centile). Of 21,795 participants, 6,722 were excluded, leaving 15,073. Small differences were observed between the first and best trials (mean difference in FEV1: 0.049 L [95% CI: 0.049, 0.051]; FVC: 0.053 L [95% CI: 0.052, 0.054]; FEV1/FVC: 0.0024 [95% CI: 0.002, 0.003]). Strong agreement existed between the first and best trials in ruling out obstructive impairments (99.7% sensitivity). However, the first trial falsely classified 16.6% as obstructed, posing diagnostic challenges. A simplified protocol could reduce testing time by 30%, saving 40 hours monthly. The first acceptable trial offers potential as a screening tool to rule out obstructive impairment, while full assessment is needed to confirm it. A higher LLN (e.g. 10th centile) could minimise false negatives.
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 | 0.091 | 0.253 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.005 | 0.002 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.003 | 0.005 |
| Open science | 0.003 | 0.002 |
| Research integrity | 0.003 | 0.003 |
| Insufficient payload (model declined to judge) | 0.001 | 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".