SU-E-T-92: Creation of a Comprehensive Head and Model Using Knowledge Based Planning
Bibliographic record
Abstract
Purpose: The purpose of this study was to evaluate if a commercial implementation of Knowledge Based Planning (KBP) software can be used to estimate Dose Volume Histogram (DVHs) on a comprehensive data set of Head and Neck (HN) patients. Methods: KBP is a tool capable of estimating DVHs for Organs At Risk (OARs) based on the DVHs of plans of similar patients treated in the past. This study used a newly developed commercial implementation of KBP to create a HN model. The model was trained using a database of retrospectively treated HN patients. This database covered the spectrum of cases expected to be found in the clinic, including multiple targets and 18 different dose prescription combinations. A set of independent validation patients was used to quantify the accuracy of DVHs estimated using the model and covered the same spectrum of HN cases. Results: The accuracy of the model was calculated by comparing the volumes of the estimated and clinical DVHs at doses equal to 50%, 85% and 99% of the maximum OAR dose. This allowed us to quantify the accuracy of the estimated DVHs even in cases when the OAR was receiving a low dose. The highest accuracy was obtained in the estimation of the DVHs for the Brain (<1% on average). The accuracy for the Brainstem, Cord, Mandible, Oral Cavity, Parotids and the Pharyngeal Constrictor ranged from -2% to 9% on average. Conclusion: This study shows the feasibility to estimate DVHs in a wide range of HN cases using a novel KBP algorithm. The use of a comprehensive set of patients results in a robust HN model which can be used in a wide range of clinical cases. Further planning is required to confirm if the current accuracy is sufficient to guide the planning process. The authors are Clinical Evaluators/Consultants for Varian Medical Systems. This study was partially funded by Varian Medical Systems.
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 | 0.001 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.004 | 0.001 |
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".