arXiv cs.AI· Alda Jo\~ao Andrade, M\'onica Martins, Andr\'e Ferreira, Tarc\'isio Ara\'ujo, Lu\'is Lopes, Victor Alves·· 3 小时前AI 评分23
精选 ERCP 内窥镜影像数据集发布，助力胆胰疾病 AI 诊断
Curated endoscopic retrograde cholangiopancreatography images dataset
AI 导读
研究者发布一个经整理的 ERCP（内镜逆行胰胆管造影）影像数据集，包含 19,018 张原始图像和 19,317 张处理后图像，来自 1,602 名患者，其中 5,519 张已标注。所有图像由两名 5 年以上经验的消化科医生人工检查标注，并由一名 20 年以上经验的医生复核。数据集的有效性通过分类实验验证，旨在为胆胰疾病的自动分析与诊断提供基准。
正文 · 原文
Abstract:Endoscopic Retrograde Cholangiopancreatography (ERCP) is a key procedure in the diagnosis and treatment of biliary and pancreatic diseases. Artificial intelligence has been pointed as one solution to automatize diagnosis. However, public ERCP datasets are scarce, which limits the use of such approach. Therefore, this study aims to help fill this gap by providing a large and curated dataset. The collection is composed of 19.018 raw images and 19.317 processed from 1.602 patients. 5.519 images are labeled, which provides a ready to use dataset. All images were manually inspected and annotated by two gastroenterologist with more than 5 years of experience and reviewed by another gastroenterologist with more than 20 years of experience, all with more than 400 ERCP procedures annually. The utility and validity of the dataset is proven by a classification experiment. This collection aims to provide or contribute for a benchmark in automatic ERCP analysis and diagnosis of biliary and pancreatic diseases.
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) |
| ACM classes: | E.m; I.4; I.2 |
| Cite as: | arXiv:2601.16759 [cs.CV] |
| (or arXiv:2601.16759v2 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2601.16759 arXiv-issued DOI via DataCite |
|
| Journal reference: | Sci Data 13, 1485 (2026) |
| Related DOI: | https://doi.org/10.1038/s41597-026-07679-1
DOI(s) linking to related resources |
Submission history
From: Mónica Martins Carina Costa [view email]
[v1]
Fri, 23 Jan 2026 14:05:42 UTC (1,286 KB)
[v2]
Mon, 9 Nov 2026 21:09:20 UTC (1,286 KB)
来源：arXiv cs.AI · arxiv.org