Poster ‐ 37: Pre‐clinical geometric, dosimetric and timing assessment of head and neck OARs using an in‐house atlas‐based auto‐segmentation (ABAS) tool
Bibliographic record
Abstract
Purpose: This study aims to validate the geometric and dosimetric performance of our in‐house ABAS tool using manual inter‐observer variation as benchmark data. Materials and methods: An in‐house ABAS constructed, using MIM MaestroTM version 6.5, from 36 previously treated head and neck cases. 15 OARs of eight nasopharynx patients were segmented via three observers from the same institution and using the in‐house ABAS. Percentage of volume differences (ΔV%,), degree of overlap (DICE), distance‐to‐agreement and the standard deviation of absolute dose difference ΔD SD(Gy) among the eight cases [ΔDmax SD(Gy) for serial organs and ΔDmean SD(Gy) for parallel organs] were compared between the manual segmentation and original ABAS contours for each OAR. Results: The geometric results indicated that ABAS ΔV% was within 1SD from the manual segmentation. DICE showed that manual segmentation marginally outperformed ABAS in the delineation of parotid glands, submandibular glands, and laryngopharynx, but ABSA performed as well as the manual segmentation or better for all other structures. The distance‐to‐agreement was <1.5 cm for 87% of ABAS structures. From a dosimetric perspective, only ABAS salivary glands demonstrated higher ΔD SD(Gy) compared to manual segmentation. For all other structures ABAS results were similar or better than the manual segmentation. The average time to segment a complete H&N OAR set was <3minutes versus 30minutes for manual segmentation. Conclusion: ABAS is a practical time‐saving tool. This study indicated up 90% time‐saving of the operator time per case. ABAS geometric and dosimetric results were within ±1SD inter‐observer variation for most of the structures.
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 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.000 |
| Insufficient payload (model declined to judge) | 0.010 | 0.003 |
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".