SU‐E‐J‐42: Customized Deformable Image Registration Using Open‐Source Software SlicerRT
Bibliographic record
Abstract
Purpose: SlicerRT is a flexible platform that allows the user to incorporate the necessary images registration and processing tools to improve clinical workflow. This work validates the accuracy and the versatility of the deformable image registration algorithm of the free open‐source software SlicerRT using a deformable physical pelvic phantom versus available commercial image fusion algorithms. Methods: Optical camera images of nonradiopaque markers implanted in an anatomical pelvic phantom were used to measure the ground‐truth deformation and evaluate the theoretical deformations for several DIR algorithms. To perform the registration, full and empty bladder computed tomography (CT) images of the phantom were obtained and used as fixed and moving images, respectively. The DIR module, found in SlicerRT, used a B‐spline deformable image registration with multiple optimization parameters that allowed customization of the registration including a regularization term that controlled the amount of local voxel displacement. The virtual deformation field at the center of the phantom was obtained and compared to the experimental ground‐truth values. The parameters of SlicerRT were then varied to improve spatial accuracy. To quantify image similarity, the mean absolute difference (MAD) parameter using Hounsfield units was calculated. In addition, the Dice coefficient of the contoured rectum was evaluated to validate the strength of the algorithm to transfer anatomical contours. Results: Overall, SlicerRT achieved one of the lowest MAD values across the algorithm spectrum, but slightly smaller mean spatial errors in comparison to MIM software (MIM). On the other hand, SlicerRT created higher mean spatial errors than Velocity Medical Solutions (VEL), although obtaining an improvement on the DICE to 0.91. The large spatial errors were attributed to the poor contrast in the prostate bladder interface of the phantom. Conclusion: Based phantom validation, SlicerRT is capable of achieving comparable DIR accuracy to commercial programs such as MIM and VEL.
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 | 0.002 | 0.005 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.014 | 0.004 |
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".