Assessment and improvement of radiation oncology trainee contouring ability utilizing consensus‐based penalty metrics
Bibliographic record
Abstract
INTRODUCTION: The objective of this study was to develop and assess the feasibility of utilizing consensus-based penalty metrics for the purpose of critical structure and organ at risk (OAR) contouring quality assurance and improvement. METHODS: A Delphi study was conducted to obtain consensus on contouring penalty metrics to assess trainee-generated OAR contours. Voxel-based penalty metric equations were used to score regions of discordance between trainee and expert contour sets. The utility of these penalty metric scores for objective feedback on contouring quality was assessed by using cases prepared for weekly radiation oncology radiation oncology trainee treatment planning rounds. RESULTS: In two Delphi rounds, six radiation oncology specialists reached agreement on clinical importance/impact and organ radiosensitivity as the two primary criteria for the creation of the Critical Structure Inter-comparison of Segmentation (CriSIS) penalty functions. Linear/quadratic penalty scoring functions (for over- and under-contouring) with one of four levels of severity (none, low, moderate and high) were assigned for each of 20 OARs in order to generate a CriSIS score when new OAR contours are compared with reference/expert standards. Six cases (central nervous system, head and neck, gastrointestinal, genitourinary, gynaecological and thoracic) then were used to validate 18 OAR metrics through comparison of trainee and expert contour sets using the consensus derived CriSIS functions. For 14 OARs, there was an improvement in CriSIS score post-educational intervention. CONCLUSIONS: The use of consensus-based contouring penalty metrics to provide quantitative information for contouring improvement is feasible.
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 | 0.003 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 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 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".