SU‐E‐T‐274: Evaluation of Atlas‐Based Segmentation Algorithms: VelocityAI vs. MIMvista
Bibliographic record
Abstract
Purpose: IMRT is driven by volumetric segmentation, thus greater care and accuracy are necessary when contouring structures. The contouring process requires staff experience and ample time. We have evaluated the performance of two commercial atlas‐based segmentation algorithms. Methods: VelocityAI and MIMvista were compared. Twenty‐one IMRT head and neck cases were randomly and retrospectively chosen. These cases included their respective CT scans and physician‐drawn structures. The twenty‐one cases were divided into two sets: one to create the atlas (eleven) and the other to test the atlas on (ten). In MIMvista the atlas was created using the in‐software tool and setting the most representative patient as the template. In VelocityAI the atlas was created using all ten cases to create an average patient atlas. The averaging used on the ten cases for the VelocityAI atlas was created using the STAPLE algorithm (provided by Velocity Medical Solutions).Results: Twelve OARs were compared to physician‐drawn structures using the Dice similarity coefficient (DSC). VelocityAI and MIMvista performed quite well on the brain, brainstem, spinal cord and eyes with mean DSCs ranging between 0.770– 0.947 for VelocityAI and 0.647– 0.978 for MIMvista. Neither program performed too well on the esophagus, larynx, oral cavity, parotids and sphincter muscle with mean DSCs ranging between 0.348– 0.690 for VelocityAI and 0.389–0.709 for MIMvista Conclusions: This work revealed that neither of the software truly outperformed the other. MIMvista did yield slightly better structures, yet the problem is that much modification is required to render the structures valid. VelocityAI and MIMvista both have great potential and can be used to provide a quick draft of structures, which may reduce physician‐contouring time.
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 | 0.001 | 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) | 0.001 | 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".