Penn Medicine BioBank (PMBB)
收藏arXiv2025-09-30 收录
下载链接：
https://github.com/allisonjchae/DMT2RiskAssessment
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含了来自宾夕法尼亚大学健康系统的超过6万名患者的数据，这些数据与超过1万个ICD-9编码相关联，重点关注了去识别化的患者人口统计学和临床变量，包括HbA1c测量值。在最终筛选后的数据集中，共有389名患者，进行了2077次HbA1c测量，这些数据按照自我报告的性别、种族和年龄进行了分层。数据集还包括了临床实验室数值和基于图像的表型。规模上，该数据集覆盖了超过6万名患者，任务是对糖尿病风险进行分层。
This dataset comprises data from over 60,000 patients associated with the University of Pennsylvania Health System, linked to more than 10,000 ICD-9 codes. It centers on de-identified patient demographic and clinical variables, including HbA1c measurements. Following final filtering, the curated dataset includes 389 patients with a total of 2,077 HbA1c measurements, which are stratified by self-reported gender, race, and age. Additionally, the dataset contains clinical laboratory values and image-based phenotyping. Overall, this dataset covers over 60,000 patients, with its primary task being diabetes risk stratification.
提供机构：
University of Pennsylvania
