PALM
收藏资源简介:
PALM数据集是由中山大学中山眼科中心与百度智能医疗部门合作创建的开放式眼底图像数据集,专注于病理性近视(PM)的识别与解剖结构标注。该数据集包含1200张彩色眼底照片,每张照片均附有病理性近视类别标签及视盘、黄斑位置和病变如斑片状视网膜萎缩(包括视乳头周围萎缩)和视网膜脱离的手动标注。数据集的创建过程涉及详细的标注流程,确保样本质量和特征的准确性。PALM数据集的应用领域主要集中在通过人工智能方法辅助临床医生进行疾病标志的识别或大规模人群筛查,以早期诊断PM,预防视力损失。
The PALM dataset is an open-access fundus image dataset jointly developed by the Zhongshan Ophthalmic Center of Sun Yat-sen University and Baidu's Intelligent Medical Department, focusing on pathological myopia (PM) identification and anatomical structure annotation. It comprises 1200 color fundus photographs, each paired with pathological myopia category labels, alongside manual annotations for the positions of the optic disc and macula, as well as lesions including patchy retinal atrophy (specifically peripapillary atrophy) and retinal detachment. A rigorous annotation workflow was implemented during the dataset's development to ensure sample quality and the accuracy of annotated features. The primary applications of the PALM dataset center on assisting clinicians in disease sign identification or conducting large-scale population screening via artificial intelligence approaches, to achieve early diagnosis of PM and prevent vision loss.

- 1Palm: A Culturally Inclusive and Linguistically Diverse Dataset for Arabic LLMs不列颠哥伦比亚大学、MBZUAI、Invertible AI、伯泽特大学、沙特王子大学、UM6P、开罗大学、 Ain Shams大学、大马士革大学、苏丹大学、曼努菲亚大学、努瓦克肖特大学、阿尔及利亚国立理工学院、全帆大学、阿尔法塞尔大学 · 2025年



