Application of Risk Prediction Models to Lung Cancer Screening
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Globally, lung cancer is the leading cause of cancer death and is a major public health problem. Because lung cancer is usually diagnosed at an advanced stage, survival is generally poor. In recent decades, clinical advances have not led to marked improvements in outcomes. A recent advance of importance arose when the National Lung Screening Trial (NLST) findings indicated that low-dose computed tomography screening of high-risk individuals can lead to a lung cancer mortality reduction of 20%. NLST identified high-risk individuals using the following criteria: age 55 to 74 years; ≥30 pack-years of smoking; and number of years since smoking cessation ≤15 years. Medical screening is most effective when applied to high-risk individuals. The NLST criteria for high risk were practical for enrolling individuals into a clinical trial but are not optimal for risk estimation. Lung cancer risk prediction models are expected to be superior. Indeed, recently, 3 studies have provided quantitative evidence that selection of individuals for lung screening on the basis of estimates from high-quality risk prediction models is superior to using NLST criteria or similar criteria, such as the United States Preventive Services Task Force (USPSTF) criteria. Compared with NLST/USPSTF criteria, selection of individuals for screening using high-quality risk models should lead to fewer individuals being screened, more cancers being detected, and fewer false positives. More lives will be saved with greater cost-effectiveness. In this paper, we review methodological background for prediction modeling, existing lung cancer risk prediction models and some of their findings, and current issues in lung cancer risk prediction modeling and discuss future research.
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.
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 | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.002 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 it