糖尿病专病数据集

电子病历文本细项解析方法：首先对于病历文本数据，进行分层解析，将获取到的文本数据按事件流的方式进行拆解，根据需要解析的各项信息，有针对性的选取包含该信息的内容类别事件流，进一步行各项指标的细项结构化解析。 多维诊疗数据构建患者主索引：将患者数据特征向量定义为患者性别、住址、家族遗传病、过敏原等信息，使用DBSCAN算法，基于特征向量的密度，将密度相近的数据点划为同一个簇，将患者数据点进行聚类，每个聚类可以视为一个患者群体，作为主索引的标识。Method for Fine-grained Parsing of Electronic Medical Record (EMR) Texts: Firstly, hierarchical parsing is conducted on medical record text data. The acquired text data is decomposed into event streams. According to the specific information items to be parsed, the event streams of content categories containing the target information are selected in a targeted manner, followed by further fine-grained structured parsing of various indicators. Construction of Patient Master Index (PMI) from Multi-dimensional Diagnosis and Treatment Data: The feature vector of patient data is defined as information including patient gender, address, family genetic diseases, allergens and other relevant details. The DBSCAN algorithm is employed to cluster patient data points based on the density of their feature vectors, grouping data points with similar densities into the same cluster. Each cluster can be regarded as a patient group, which acts as the identifier for the master index.

搜集汇总

数据集介绍

特点

糖尿病专病数据集是一个包含15亿条数据的大型医疗数据集，每月更新，涵盖就诊信息、诊断疾病、药品、检验检查等多个方面，适用于医疗、教学和科研领域的诊疗模式研究和药物经济学研究。