Quantitative PET lesion delineation using a three-dimensional adaptive threshold region growing algorithm (RGA): A method for calibration
Bibliographic record
Abstract
1443 Objectives The goals of this study were to: (1) investigate the dependence of lesion size and tumor-to-background ratio (TBR) on the percent input parameter of RGA-based PET lesion delineations, and (2) develop and evaluate objective and quantitative methods for calibrating the RGA percent input parameter. Methods Five uniform-activity spherical phantom lesions (0.54 mL – 11.58 mL) of 11C within a background of 18F were imaged on a PET/CT scanner for two hours in 3D dynamic list mode yielding TBRs from 1.2 to 30. Images were reconstructed at two minute intervals using a clinical RAMLA 3D algorithm with CT attenuation correction. The RGA percent input parameter corresponding to a sharp increase in delineated volume was observed [Li et al. Med Phys. 2008;35:3711–3721] and employed as a calibration feature. Delineations using the proposed calibration method were compared to those obtained through application of four fixed RGA input parameters (from 55% to 85%). Results The percent input parameter required for the RGA to yield correct delineations was found to decrease both as a function of increasing sphere volume and increasing TBR. Delineations obtained using the proposed calibration were found to consistently outperform those obtained with fixed input parameters. In addition, the average sum of voxel intensities for the calibration-based delineations was found to be linearly correlated with the injected lesion activity (R2 = 0.998 for TBR > 5). Conclusions An objectively identified lesion-specific image feature was successfully implemented for calibration of the percent input parameter for RGA-based delineations.
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 | 0.008 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.002 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 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".