P.103 Automated pituitary adenoma segmentation for radiosurgery with deep learning-based model
Bibliographic record
Abstract
Background: Pituitary adenomas are treated with endoscopic surgery, while stereotactic radiosurgery addresses complex cases. Our study highlights AI’s role in accurate segmentation, improving treatment planning workflow efficiency Methods: In a retrospective study at Na Homolce Hospital (January 2010 to October 2022), SRS for pituitary adenomas was analyzed. Data were split 80:20 for training and validation. Using nnU-net, a medical image segmentation tool, a model predicted precise tumor, optic nerve, and pituitary gland segmentation. Accuracy was evaluated quantitatively with Dice similarity coefficient and qualitatively by human experts. The study explored the impact of tumor volume and hormonal activity status on segmentation accuracy. Results: The study comprised 582 and 146 patients in training and validation sets, respectively. The model achieved Dice similarity coefficients of 83.1% (tumor), 62.9% (normal gland), and 78.0% (optic nerve). Expert assessments deemed 41% directly applicable, 31.5% needing minor adjustments, and 27.4% unsuitable for clinical use. Larger tumor volume and non-functioning adenomas correlated with higher accuracy. Including T2 weighted scans improved DSC for optic nerve and normal gland. Conclusions: The study showcases deep learning’s potential in automating pituitary adenoma segmentation from MRI data, particularly excelling in large, hormonally inactive macroadenomas. Encourages collaborative use with clinicians for improved neurosurgical patient care.
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 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 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".