AI Science Tracker
EN

A stenosis detector was tested on an external coronary angiography cohort

University of Oxford; Oxford University Hospitals NHS Trust · 23 Oct 2026 · medicine

Discovery

A stenosis detector was tested on an external coronary angiography cohort

Source check: Source checked. Not independent scientific confirmation.

Finding: A learned detector using coronary-vessel features achieved AP@50 of 0.434 on 48 angiography images from 27 Oxford Acute Myocardial Infarction study patients. Test-time augmentation raised it to 0.480. Standard YOLOv11-L scored 0.339 without augmentation. These are lesion-localization scores from a small retrospective cohort.

Technology: Fusion-YOLOv11-L; UNet++; learned vessel segmentation and lesion detection

Publication: Preprint · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.

AI contribution
Fusion-YOLOv11-L combined learned coronary-vessel segmentation features and lesion detection features to localize stenoses in angiography frames.
Test and result
The study trained and tuned the model on ARCADE-derived data and assessed localization of 85 stenoses on separately sourced OxAMI frames.
Limitation / still unknown
The result covers selected end-diastolic frames and a small external sample. Patient-level diagnostic performance and clinical outcomes were outside this evaluation.
Evidence sources and annotation context

Evidence context (descriptions follow the cited sources; test type is a catalog classification):

Evidence sources: https://arxiv.org/abs/2610.32730v1 · https://arxiv.org/html/2610.32730v1

Source checked · editorial details

The source page was checked against this entry. Independent confirmation is not assessed by this label.

Source type: paper. Paper: linked separately.

Review date: 21 Nov 2026. Review scope: arXiv v1 date, Methods §2.1–2.4 and external test results, especially Table 5.

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 2/5 · Validation 2/5 · Novelty 2/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.

Open in the full catalog for filters, charts and related discoveries. Browse medicine discoveries.

Related catalog entries

Links use documented relationships where available, plus shared catalog context. A shared topic or model family does not establish a scientific dependency.