Poster — Thur Eve — 43: Clinical Evaluation of an Atlas Based Auto‐Segmentation Application for Auto‐Contouring Pelvic Targets and OARs
Bibliographic record
Abstract
The use of volumetric image data sets in radiotherapy treatment planning has become standard practice and requires an accurate delineation of anatomical structures, a tedious and time consuming task. This has lead to the development of auto‐segmentation approaches for contouring anatomical structures on CT data sets. In this work, we evaluate the accuracy of the ABAS software package (Atlas Based Auto‐Segmentation, version 1.1.0, CMS, Inc, The Elekta Group) for contouring pelvic organs relevant to radiotherapy. The evaluation is based on randomly selected CT data sets of 8 patients treated in our clinic for pelvic malignancies, in which relevant organs have been manually contoured by an oncologist. Each of the 8 CT data sets is used as an atlas to generate contours on all other sets. The resulting ABAS‐generated contours are compared to oncologists ones by means of the Dice coefficient. We find that contours generated by ABAS are best and will require no editing for bony structures such as the femoral heads and the pelvis. For the prostate, bladder and the rectum, ABAS contours are acceptable but will require some editing. However, for the prostate, the presence of fiducial implants could lead to streaking artifacts that degrade the quality of ABAS‐generated contours. For smaller organs with low tissue contrasts such as seminal vesicles and iliac nodes, ABAS results are poor and will lead to no time savings in the radiotherapy process. Finally for organs for which ABAS works well it is quite robust against the variability of the atlas.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.000 |
| 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 | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 teacher head, 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".