Differentiating clinically important interstitial lung abnormalities in lung cancer screening
Bibliographic record
Abstract
BACKGROUND: Interstitial lung abnormalities (ILAs) are common incidental findings in lung cancer screening (LCS). However, challenges remain in identifying clinically relevant ILAs as highlighted in a joint statement by a European multidisciplinary task force led by the European Respiratory Society (ERS). To address these challenges, we analysed ILAs identified in one of Europe's largest LCS studies. METHODS: Of 11 635 LCS individuals, 417 screen-detected ILAs were evaluated using a new visual classification system focused on traction bronchiolectasis: non-fibrotic ILA (no traction bronchiolectasis), fibrotic ILA (traction bronchiolectasis in ≤2 lobes); undiagnosed interstitial lung disease (traction bronchiolectasis in >2 lobes). Observer agreement was compared with Fleischner Society ILA classification using Cohen's Kappa. An age, sex and smoking history-matched control group allowed the examination of associations between baseline ILA/UILD and comorbidities, forced vital capacity (FVC), hospitalisations (Student's t-tests) and mortality (univariable and multivariable Cox proportional hazards models). FINDINGS: Our visual ILA classification showed superior interobserver agreement (K=0.76) versus the Fleischner ILA classification (K=0.64). ILA/UILD subjects had more prevalent comorbidities, increasing (vs controls) approximately 10 years prior to ILA/UILD diagnosis. Compared with controls, mortality rates were 6-fold higher for UILD participants and 3-fold higher for fibrotic and non-fibrotic ILA subtypes. On multivariable Cox regression analysis, ILA/UILD presence (HR=4.90, 95% CI =2.36 to 10.10, p<0.001) showed stronger independent associations with mortality than baseline FVC (HR=0.98, 95% CI =0.96 to 1.00, p=0.04). CONCLUSION: We demonstrate a new reproducible classification of clinically important ILA/UILDs in LCS populations. We highlight that FVC shows limited associations with mortality in ILA/UILD subjects. Increased multiorgan comorbidity in ILA/UILD subjects highlights a need for comprehensive early multisystem evaluation.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.006 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
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 teacher head, 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".