2026-10-11T07:42:50Z https://www.scidb.cn/oai

10.57760/sciencedb.19700 2025-06-12T13:53:35Z

2025-06-12 Comprehensive Assessment of Chronic Gastritis on WSI Data doi:10.57760/sciencedb.19700 en The dataset contains 800 whole slide images (WSIs on 40x, 600 from gastric antrum and 200 from gastric corpus) for assessing "Inflammation (Task 1)", "Activity (Task 2)", "Atrophy (Task 3)" and "Intestinal Metaplasia (Task 4)" grade of chronic gastritis (CG) without identified information, which are individually classified into "Normal", "Mild", "Moderate", "Severe" based on patients' histological alterations according to the updated Sydney System. Given the challenges of biopsying completely normal gastric mucosa, devoid of chronic inflammation cells and neutrophils, in real-world clinical settings, our dataset did not encompass such samples. Therefore, task1 and task2 are additionally classified into "Normal to Mild", "Moderate", "Severe". Please see https://github.com/nicedoctor123/Gastritis-MIL-pathology for the details on dealing with these WSIs. We request that you cite our work published in Patterns when using this dataset. Deep learning; Chronic Gastritis; whole slide images; WSI Kun Xia Yihuang Hu Liangsheng Wang Hongzhi Xu PUBLIC https://creativecommons.org/licenses/by-nc-sa/4.0/ dataset http://www.doi.org/10.1016/j.patter.2025.101286 Science Data Bank