SU-E-J-41: Quantitative Assessment of Anatomical Changes throughout the Course Radiation Therapy with Deformable Registration
Bibliographic record
Abstract
Introduction: Using Image guided radiation therapy on a regular basis can increase the clinical burden substantially. Our goal is to develop semi-automated deformable registration tools to monitor anatomical changes in order to make consistent treatment decisions without increasing the clinical workload. Methods: We used a B-spline deformable image registration package to analyze images obtained from IGRT procedures for lung cancer patients. CT and CBCT datasets were imported into our software and processed as follow: (1) physician contours were extracted via the DICOM-RT protocol; (2) A first deformable registration was performed to register the plan dataset to the first CBCT and deformation maps were applied to contours to generate new contours for the CBCTs; (3) deformable registrations were then applied between all CBCTs and new set of contours adapted to the changed anatomy were automatically generated. Information extracted from each registration was: average magnitude of deformation; displacement of the center-of-mass; regions most affected by deformations. Results: Deformations were performed on 40 lung CBCTs. A visual inspection of each case found no significant anatomical errors in the registration although in case of strong deformation the deformation map slightly underestimated anatomical changes. Precision was assed by repeating the same deformable registration 10 times. We found variations of less than 0.1 mm (1 standard deviation) thus indicating that the deformation process is precise. The average magnitude of the displacement vector and the variation in the center-of-mass of each contour varied between 0.9 mm and 16 mm. We could differentiate between displacement of a contour and anatomical variation by taking the ratio between these two values. An action threshold of 6 mm for both was used to differentiate between “strong” anatomical change and “small” anatomical changes. Conclusion: We have developed a semi-automated deformation registration tool that let us consistently monitor anatomical changes on weekly/daily CBCTs
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 | 0.003 | 0.005 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.000 |
| 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".