Using an Artificial Intelligence System “Limbus AI” for Automatic Segmentation of OAR's in Radiotherapy: Evaluation of Effectiveness and Dosimetric Impact
Bibliographic record
Abstract
Accurate definition of organs at risk and target volumes is a critically important step in the radiotherapy workflow prior to therapy delivery. This activity is time-consuming for the operator, who performs it manually. Self-contouring software based on deep learning has been developed to improve efficiency; in this study, we examined the commercial software Limbus Contour (LC), version 1.5.0 (Limbus AI Inc., Regina, SK, Canada), which exploits deep convolutional neural networks, evaluating its impact on the radiotherapy workflow in three specific pathological sites: head-neck, left breast, and prostate. We compared manual versus automated segmentation on selected organs at risk associated with specific types of treatment, evaluating geometric performance (volume change, Dice Index), dosimetric impact (differences in DVHs), and timing of the two contouring modes. The average Dice Index value obtained was 84%; in particular, the best results were recorded for the lungs, but equally good results were obtained for the bladder, femoral heads, and heart. The greatest time saving, on the other hand, was recorded in head-neck tumors, with a saving of about 80 minutes. Evaluating DVHs, the greatest dosimetric difference was found for prostate cancer in patient 1, where the penile bulb showed an average dose difference, in the two contouring modes, of 73%. The contours and data generated by AI were compared and validated by oncologists in our department, who confirmed that the use of Limbus Contour helped reduce the inter-operator discrepancies typical of manual contouring. The introduction of this self-contouring software brought an improvement of workflow in Rieti radiotherapy department by significantly reducing procedure time, maintaining high quality of treatments, and reducing variability between operators.
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 | 0.003 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| 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".