TU‐F‐17A‐03: A 4D Lung Phantom for Coupled Registration/Segmentation Evaluation
Bibliographic record
Abstract
Purpose: Coupling the processes of segmentation and registration (regmentation) is a recent development that allows improved efficiency and accuracy for both steps and may improve the clinical feasibility of online adaptive radiotherapy. Presented is a multimodality animal tissue model designed specifically to provide a ground truth to simultaneously evaluate segmentation and registration errors during respiratory motion. Methods: Tumor surrogates were constructed from vacuum sealed hydrated natural sea sponges with catheters used for the injection of PET radiotracer. These contained two compartments allowing for two concentrations of radiotracer mimicking both tumor and background signals. The lungs were inflated to different volumes using an air pump and flow valve and scanned using PET/CT and MRI. Anatomical landmarks were used to evaluate the registration accuracy using an automated bifurcation tracking pipeline for reproducibility. The bifurcation tracking accuracy was assessed using virtual deformations of 2.6 cm, 5.2 cm and 7.8 cm of a CT scan of a corresponding human thorax. Bifurcations were detected in the deformed dataset and compared to known deformation coordinates for 76 points. Results: The bifurcation tracking accuracy was found to have a mean error of −0.94, 0.79 and −0.57 voxels in the left‐right, anterior‐posterior and inferior‐superior axes using a 1×1×5 mm3 resolution after the CT volume was deformed 7.8 cm. The tumor surrogates provided a segmentation ground truth after being registered to the phantom image. Conclusion: A swine lung model in conjunction with vacuum sealed sponges and a bifurcation tracking algorithm is presented that is MRI, PET and CT compatible and anatomically and kinetically realistic. Corresponding software for tracking anatomical landmarks within the phantom shows sub‐voxel accuracy. Vacuum sealed sponges provide realistic tumor surrogate with a known boundary. A ground truth with minimal uncertainty is thus realized that can be used for comparing the performance of registration and segmentation algorithms.
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) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.000 |
| 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".