Bibliographic record
Abstract
Introduction: Current prognostic tools in non-muscle-invasive bladder cancer (nmIBC) perform poorly and do not fully reflect contemporary practices.We aimed to develop, externally validate, and conduct an algorithmic audit of a progression risk assessment tool using artificial intelligence approaches (progrxn-BCa).Methods: progrxn-BCa, based on a random survival forest, was trained on nmIBC patients treated from January 1, 2005, to June 30, 2022, at four Canadian academic or community hospitals.external validation was performed on patients treated from november 1, 2011, to september 11, 2023, across 13 institutions from the Canadian Bladder Cancer information system.the primary outcome was time to progression, defined as first development of muscle-invasive or metastatic disease.progrxn-BCa was compared to the european Association of Urology risk calculator and a lAsso Cox model using identical variables as progrxn-BCa.model performance in predicting five-year progression risk was characterized using c-index, calibration plots, decision curve analysis, and an algorithmic audit.Results: overall, 999 of 7032 patients (14%) developed progression during a median followup of 3.0 years (IQr 1.4-5.4).progrxn-BCa had the highest c-index overall (training: 0.83, 95% CI 0.81-0.84;validation: 0.76, 95% CI 0.74-0.77)and across different subgroups.It was well-calibrated and had the highest net benefit for clinically relevant thresholds from 15-40%.False negatives occurred in only 2-6% of all predictions, most commonly found in patients with ta disease.progrxn-BCa could better substratify intermediate-risk patients compared to current guideline recommendations, reclassifying 12% of these patients with an observed five-year progression risk of 31.6% who otherwise would not have been considered for treatment intensification or clinical trial enrollment (Figure 1).Conclusions: progrxn-BCa outperformed current prognostication tools and improved substratification of the heterogenous intermediate-risk group.
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 | 0.002 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.003 | 0.002 |
| Open science | 0.002 | 0.004 |
| Research integrity | 0.003 | 0.004 |
| Insufficient payload (model declined to judge) | 0.310 | 0.148 |
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; the direct Gemma label and the distilled Codex classifier agree on what is shown here.
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".