Multi-Class Segmentation of Aortic Branches and Zones on Computed Tomography Angiography
Bibliographic record
Abstract
The aorta, the body's largest artery, can face potential threats like dissection and aneurysm, requiring prompt surgical intervention. Traditional surgical techniques for aortic disease often carry significant risks. Recent advancements in medical imaging, particularly computed tomography angiography (CTA), and minimally invasive approaches like endovascular grafting, offer a promising alternative. Accurate 3D segmentation of the aorta and its branches and zones on CTA is crucial for successful interventions. Inaccurate segmentation can lead to critical errors in surgical planning and endograft design, jeopardizing patient safety and treatment outcomes. While machine learning has revolutionized 3D medical image analysis, its potential in acute uncomplicated type B aortic dissection (auTBAD), the most common aortic emergency, remains largely unexplored. In the clinical realm, auTBAD is sub-categorized using SVS/STS zones, a detailed classification system defined by specific zones of the aorta in relation to aortic branches. Surgeons rely on this classification system to determine the optimum treatment algorithm for each patient. Current methods for aortic segmentation often treat it as a binary segmentation problem, neglecting the essential differentiation between individual aortic branches and their relationships to SVS/STS zones. This challenge addresses these limitations by offering the first large-scale dataset of 100 CTA volumes paired with detailed annotations for 23 different aortic branches and the clinically relevant SVS/STS zones. Participating teams will have the opportunity to develop innovative algorithms for accurate, automated, and multi-class segmentation of this intricate vascular structure. By fostering advancements in image analysis techniques for CTA, this challenge aims to:(1) Improve clinical care for patients with aortic diseases by enabling accurate diagnosis, more precise surgical planning, and potentially safer, minimally invasive interventions.(2) Bring greater attention and research focus to auTBAD, a relatively rare and challenging disease, potentially leading to novel treatment strategies.(3) Bridge interdisciplinary communication between researchers in medical image analysis, computer vision, and machine learning, paving the way for collaborative solutions to overcome technical barriers in complex aortic segmentation tasks. In summary, this challenge has three main features:(1) Task: this is the first challenge for the segmentation of aortic branches and zones in CTA scans.(2) Dataset: we provide the largest annotated dataset for aortic segmentation, including 100 3D CTA scans.(3) Evaluation: we focus on both segmentation accuracy and segmentation efficiency.
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 | 0.002 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.005 | 0.002 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.003 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.002 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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".