MFreidank/glenda
收藏资源简介:
GLENDA(妇科腹腔镜子宫内膜异位症数据集)包含超过350张注释的子宫内膜异位症病变图像,这些图像来自100多次妇科腹腔镜手术,以及来自20多次手术的超过13,000张未注释的非病理图像。该数据集专门用于自动内容分析问题,特别是在子宫内膜异位症识别的背景下。数据集的结构包括二进制分类和多类分类配置,每个配置都有详细的特征描述和数据分割信息。数据集的注释由子宫内膜异位症治疗领域的医学专家生成。数据集的使用仅限于科学研究目的,不能用于商业用途。
GLENDA (Gynecological Laparoscopic Endometriosis Dataset) contains over 350 annotated images of endometriotic lesions, collected from more than 100 gynecological laparoscopic surgeries, alongside over 13,000 unannotated non-pathological images from more than 20 additional surgeries. This dataset is specifically designed for automated content analysis tasks, particularly in the context of endometriosis detection. The dataset structure supports both binary and multi-class classification setups, with detailed feature descriptions and data splitting information provided for each configuration. All annotations in the dataset were generated by medical experts specializing in endometriosis treatment. Usage of this dataset is strictly limited to scientific research purposes, and commercial use is prohibited.
数据集概述
名称: GLENDA - The ITEC Gynecologic Laparoscopy Endometriosis Dataset
语言: 英语(en)
许可证: CC BY-NC 4.0
多语言性: 单语种
大小: 10K<n<100K
任务类别: 图像分类
数据集结构
配置名称
- binary_classification
- multiclass_classification
特征
- 图像: 数据类型为图像
- 元数据: 结构包括id, width, height, file_name, path, flickr_url, coco_url, date_captured, case_id, video_id, frame_id, from_seconds, to_seconds
- 标签: 数据类型为类标签,binary_classification包含no_pathology和endometriosis两个类别,multiclass_classification包含No-Pathology, 6.1.1.1_Endo-Peritoneum, 6.1.1.2_Endo-Ovar, 6.1.1.3_Endo-TIE, 6.1.1.4_Endo-Uterus五个类别
数据分割
- 训练集: 大小为4524957字节,包含13811个样本
下载与数据集大小
- 下载大小: 895554144字节
- 数据集大小: 4524957字节




