Le Floch Maxime, Wolf Fabian, McIntyre Lucian, Weinert Christoph, Palm Albrecht, Volk Konrad, Herzog Paul, Kirk Sophie Helene, Steinhäuser Jonas L, Stopp Catrein, Geissler Mark Enrik, Herzog Moritz, Sulk Stefan, Kather Jakob Nikolas, Meining Alexander, Hann Alexander, Palm Steffen, Hampe Jochen, Herzog Nora, Brinkmann Franz
Else Kröner Fresenius Center for Digital Health, Technische Universität Dresden (TU Dresden), Dresden, Germany.
Department of Medicine I, University Hospital Dresden, Technische Universität Dresden (TU Dresden), Dresden, Germany.
Sci Data. 2025 May 20;12(1):828. doi: 10.1038/s41597-025-05112-7.
Video capsule endoscopy (VCE) is an important technology with many advantages (non-invasive, representation of small bowel), but faces many limitations as well (time-consuming analysis, short battery lifetime, and poor image quality). Artificial intelligence (AI) holds potential to address every one of these challenges, however the progression of machine learning methods is limited by the avaibility of extensive data. We propose Galar, the most comprehensive dataset of VCE to date. Galar consists of 80 videos, culminating in 3,513,539 annotated frames covering functional, anatomical, and pathological aspects and introducing a selection of 29 distinct labels. The multisystem and multicenter VCE data from two centers in Saxony (Germany), was annotated framewise and cross-validated by five annotators. The vast scope of annotation and size of Galar make the dataset a valuable resource for the use of AI models in VCE, thereby facilitating research in diagnostic methods, patient care workflow, and the development of predictive analytics in the field.
视频胶囊内镜检查（VCE）是一项具有诸多优势（无创、可呈现小肠情况）的重要技术，但也面临许多局限（分析耗时、电池续航时间短以及图像质量差）。人工智能（AI）有潜力应对上述每一项挑战，然而机器学习方法的进展受到大量数据可用性的限制。我们提出了Galar，这是迄今为止最全面的VCE数据集。Galar由80个视频组成，最终形成了3513539个带注释的帧，涵盖功能、解剖和病理方面，并引入了29种不同标签的选择。来自德国萨克森州两个中心的多系统、多中心VCE数据逐帧进行了注释，并由五名注释者进行了交叉验证。Galar广泛的注释范围和规模使其成为VCE中使用AI模型的宝贵资源，从而促进该领域诊断方法、患者护理工作流程以及预测分析开发方面的研究。
