Computational approaches to support comparative analysis of multiparametric tests: Modelling versus Training
Bibliographic record
Abstract
Multiparametric assays for risk stratification are widely used in the management of breast cancer, with applications being developed for a number of other cancer settings. Recent data from multiple sources suggests that different tests may provide different risk estimates at the individual patient level. There is an increasing need for robust methods to support cost effective comparisons of test performance in multiple settings. The derivation of similar risk classifications using genes comprising the following multi-parametric tests Oncotype DX® (Genomic Health.), Prosigna™ (NanoString Technologies, Inc.), MammaPrint® (Agendia Inc.) was performed using different computational approaches. Results were compared to the actual test results. Two widely used approaches were applied, firstly computational "modelling" of test results using published algorithms and secondly a "training" approach which used reference results from the commercially supplied tests. We demonstrate the potential for errors to arise when using a "modelling" approach without reference to real world test results. Simultaneously we show that a "training" approach can provide a highly cost-effective solution to the development of real-world comparisons between different multigene signatures. Comparisons between existing multiparametric tests is challenging, and evidence on discordance between tests in risk stratification presents further dilemmas. We present an approach, modelled in breast cancer, which can provide health care providers and researchers with the potential to perform robust and meaningful comparisons between multigene tests in a cost-effective manner. We demonstrate that whilst viable estimates of gene signatures can be derived from modelling approaches, in our study using a training approach allowed a close approximation to true signature results.
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 | 0.007 | 0.039 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.003 | 0.002 |
| Research integrity | 0.003 | 0.003 |
| Insufficient payload (model declined to judge) | 0.003 | 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 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".