A Two-Step Optimization Method for Improving Multiple Brain Lesion Treatments with Robotic Radiosurgery
Bibliographic record
Abstract
Planning robotic radiosurgery treatments for multiple (n > 3) metastatic brain lesions is challenging due to the need of satisfying a large number of dose-volume constraints and the requirement of prescribing different dose levels to individual targets. In this study, we developed a sequential two-step optimization technique to improve the planning quality of such treatments. In contrast to the conventional approach of where all targets are simultaneously planned, we have developed a two-step optimization method. In this method, the first step was to create treatment plans for individual targets. In the second step, the 3D dose matrices associated with each plan were exported to Dicom-RT digital files and subsequently optimized. For the optimization, a singular-value-decomposition (SVD) algorithm was implemented to minimize the dose interferences among different targets. Finally, we compared the optimized treatment plans with the treatment plans created using the conventional method to determine the effectiveness of the new method. Large improvements in target dose distributions as well as normal brain sparing were found for the two-step optimization treatment plans as compared with the conventional treatment plans. The two-step optimization significantly lowered the volume of normal brain receiving relatively low doses. For example, the normal brain volume receiving 12-Gy was reduced by averaged 42% (range 34%-47%) with the two-step optimization. Such improvements generally enlarged with increasing number of targets being treated regardless of target sizes. Of note, normal brain dose was found to increase non-linearly with increasing number of targets. In summary, a two-step optimization technique is demonstrated to significantly improve the treatment plan quality as well as reduce the planning effort for multi-target robotic radiosurgery.
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 | 0.001 |
| 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 | 0.000 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 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".