X-Strata separates aneurysm sacs and overlapping vessels in two-view angiography
Source check: Source checked. Not independent scientific confirmation.
Finding: X-Strata uses learned decomposition of cine angiography and parallax between two views to infer aneurysm-sac and vessel layers and reconstruct a 3D sac surface. On 220 held-out patients from three centers, including 122 external cases, the method ranked first among 14 baselines for sac recovery in overlap regions and layer ordering. All evaluated patients had same-session 3D rotational angiography as reference; performance and clinical benefit for patients with only 2D angiography remain unestablished.
Technology: A neural network estimates angiographic layers and contrast curves; a second view supplies depth-order cues for 3D reconstruction.
Publication: Preprint · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
- AI contribution
- A neural network estimates angiographic layers and contrast curves; a second view supplies depth-order cues for 3D reconstruction.
- Test and result
- Held-out patient evaluation across three centers compared two-view X-Strata with 14 baselines and same-session 3D rotational angiography references.
- Limitation / still unknown
- The evaluated cohort had 3D rotational angiography; use in patients with only 2D imaging and any clinical-outcome benefit require separate evaluation.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://arxiv.org/html/2611.17743v1
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: paper. Paper: not linked in this catalog; this does not mean no paper exists.
Review date: 22 Nov 2026. Review scope: Primary arXiv v1 full text, including Methods, held-out evaluation, limitations, author affiliations and version date..
No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.
Editorial assessment: Impact 3/5 · Validation 2/5 · Novelty 3/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.