HS-CMU
收藏资源简介:
HS-CMU数据集由首都医科大学附属北京朝阳医院创建,专注于通过宫腔镜诊断良恶性疾病。该数据集包含175名患者的3385张高质量宫腔镜检查图像,涵盖8种不同类型的子宫内膜病变。数据采集自2023年8月至2024年1月,使用Karl Storz TC200内窥镜相机系统,分辨率为1920×1080像素。数据集创建过程中,图像由两位经验丰富的妇产科医生使用lableme软件进行标注。HS-CMU数据集旨在辅助诊断子宫内良恶性疾病,特别是在缺乏经验丰富的医生的情况下,通过深度学习算法提高诊断准确性。
HS-CMU dataset was developed by Beijing Chaoyang Hospital Affiliated to Capital Medical University, focusing on the diagnosis of benign and malignant diseases via hysteroscopy. This dataset comprises 3385 high-quality hysteroscopic images from 175 patients, covering 8 distinct types of endometrial lesions. The data was collected between August 2023 and January 2024 using the Karl Storz TC200 endoscopic camera system with a resolution of 1920×1080 pixels. During the dataset construction process, the images were annotated by two experienced obstetricians and gynecologists using the LabelMe software. The HS-CMU dataset aims to assist in the diagnosis of benign and malignant intrauterine diseases, particularly in situations where experienced clinicians are scarce, to improve diagnostic accuracy via deep learning algorithms.




