A model-based 3D patient-specific pre-treatment QA method for VMAT using the EPID
Bibliographic record
Abstract
This study reports the development and validation of a model-based, 3D patient dose reconstruction method for pre-treatment quality assurance using EPID images. The method is also investigated for sensitivity to potential MLC delivery errors. Each cine-mode EPID image acquired during plan delivery was processed using a previously developed back-projection dose reconstruction model providing a 3D dose estimate on the CT simulation data. Validation was carried out using 24 SBRT-VMAT patient plans by comparing: (1) ion chamber point dose measurements in a solid water phantom, (2) the treatment planning system (TPS) predicted 3D dose to the EPID reconstructed 3D dose in a solid water phantom, and (3) the TPS predicted 3D dose to the EPID and our forward predicted reconstructed 3D dose in the patient (CT data). AAA and AcurosXB were used for TPS predictions. Dose distributions were compared using 3%/3 mm (95% tolerance) and 2%/2 mm (90% tolerance) γ-tests in the planning target volume (PTV) and 20% dose volumes. The average percentage point dose differences between the ion chamber and the EPID, AcurosXB, and AAA were 0.73 ± 1.25%, 0.38 ± 0.96% and 1.06 ± 1.34% respectively. For the patient (CT) dose comparisons, seven (3%/3 mm) and nine (2%/2 mm) plans failed the EPID versus AAA. All plans passed the EPID versus Acuros XB and the EPID versus forward model γ-comparisons. Four types of MLC sensitive errors (opening, shifting, stuck, and retracting), of varying magnitude (0.2, 0.5, 1.0, 2.0 mm), were introduced into six different SBRT-VMAT plans. γ-comparisons of the erroneous EPID dose and original predicted dose were carried out using the same criteria as above. For all plans, the sensitivity testing using a 3%/3 mm γ-test in the PTV successfully determined MLC errors on the order of 1.0 mm, except for the single leaf retraction-type error. A 2%/2 mm criteria produced similar results with two more additional detected errors.
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 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| 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 | 0.002 |
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".