互软医疗健康数据集

本专病数据库采用结构化数据存储方式，以 UTF-8 编码的 CSV 文件为主要载体，通过唯一标识符实现表间关联。 核心数据文件涵盖患者全流程信息：大规模脱敏患者基本信息文件以病人档案 ID（唯一标识）、MPIID（主索引）为关键字，含脱敏姓名、性别等基础字段；诊断信息文件以诊断 ID 唯一标识，记录诊断名称、专病类型等信息；用药、检查检验、门诊就诊、住院治疗、手术、随访等文件均设专属唯一标识，分别记录对应诊疗关键数据；脱敏 MCI 早期筛查数据文件以筛查 ID 标识，含年龄、MMSE 评分等筛查信息。所有文件均通过病人档案 ID、MPIID 关联患者基本信息。 数据组织遵循三大原则：以 MPIID 为核心关联依据，实现跨表数据整合形成患者统一视图；按专病类型（如房颤、心衰等）分层分类，支持病种视图筛选；诊断日期等时间字段统一为 “YYYY-MM-DD HH:MM:SS” 格式，保障时序分析可用性，整体结构清晰且关联高效。This specialized disease database adopts structured data storage, with UTF-8 encoded CSV files as the primary data carrier, and establishes inter-table associations via unique identifiers. The core dataset files encompass the full-cycle information of patients: the large-scale de-identified basic patient demographic file uses patient record ID (unique identifier) and MPIID (master index) as keys, and contains basic demographic fields such as de-identified name and gender; the diagnostic information file is uniquely identified by diagnosis ID, and records diagnostic terms, specialized disease types and other relevant information; files for medication, examinations and tests, outpatient visits, inpatient treatments, surgeries, follow-up visits and other medical services are all assigned exclusive unique identifiers, and separately record corresponding key diagnosis and treatment data; the de-identified mild cognitive impairment (MCI) early screening data file is identified by screening ID, and contains screening information such as age and MMSE score. All files are linked to the basic patient demographic information via patient record ID and MPIID. The data organization follows three core principles: taking MPIID as the core association basis to achieve cross-table data integration and create a unified patient view; hierarchically classifying by specialized disease types (e.g., atrial fibrillation, heart failure, etc.) to support filtering of disease-specific views; time fields such as diagnosis date are uniformly formatted as "YYYY-MM-DD HH:MM:SS" to ensure the availability of time-series analysis, resulting in a clear overall structure and efficient inter-table associations.