Nonparametric methods for receiver operating characteristic (ROC) curve analysis in genomic studies and diagnostic medicine
Bibliographic record
Abstract
This thesis is comprised of three parts. The first part (Chapter 2) develops nonparametric statistical inference methods for partial area under ROC curves (PAUC) that can be applied to genomic studies. ROC curves are used to analyze the performance of a diagnostic test while the areas or partial areas under ROC curves are used to judge how accurately the test results can discriminate between two groups (for example, diseased and non-diseased groups). The third part of this thesis (Chapter 4) presents a novel weighted nonparametric method for estimating ROC curves. We model the probability of the disease status for a given test result by logistic regression models and we connect logistic regression and ROC curves by a weighted nonparametric method. The ROC curves fitted by this method are smoother than ROC curves produced purely by traditional nonparametric methods. More importantly, the method can be used to correct for verification bias. The second part (Chapter 3) develops methods for PAUC when results of the applicable gold standard test are incomplete, situations also referred to as data with incomplete verification or with verification bias. The true (disease) status is the 'gold standard' against which a given diagnostic test should be measured. However, there are many diseases for which the definitive diagnosis is expensive or difficult to obtain for an entire sample. We have developed a method based on the nonparametric approach for estimating partial area and its variance and tested the method by simulation studies under various situations.
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 | 0.035 | 0.135 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.003 |
| Bibliometrics | 0.006 | 0.007 |
| Science and technology studies | 0.001 | 0.004 |
| Scholarly communication | 0.004 | 0.004 |
| Open science | 0.003 | 0.003 |
| Research integrity | 0.004 | 0.007 |
| Insufficient payload (model declined to judge) | 0.005 | 0.003 |
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".