血压测量数据集

1、数据采集：通过医卫通AI一体机智能建档，在用户授权下依法合规采集用户的血压与脉搏数据。原始数据包括唯一设备标识（健康设备ID）、系统生成的匿名用户标识（健康设备监测用户ID）、舒张压（mmHg）、收缩压（mmHg）、脉搏（次/分）、测量时间戳（精确到分钟，即“血压采集时间”）及系统自动生成的状态标记（处理异常标识）。
2、数据处理：1）以“健康设备ID”和“血压采集时间”为联合主键，对原始数据进行清洗，剔除格式错误或明显超出生理范围的异常值（如收缩压<70或>250 mmHg，舒张压<40或>150 mmHg）；2）计算多维衍生指标：个体维度——日均收缩压/舒张压、血压控制达标率（<140/90 mmHg）、高血压事件频次、脉搏变异系数；设备维度——日均服务人数、设备在线率、异常上报频次；群体维度——区域高血压检出率、高脉搏人群分布；3）生成结构化、标准化的血压测量数据集，支持趋势分析、风险预警与模型训练。
3、数据应用：输出高质量的多用户血压时序数据，用于生成个人血压趋势报告、触发高血压自动预警、辅助医生调整治疗方案、评估社区高血压管理成效及训练心血管疾病风险预测AI模型。

1. Data Collection: Blood pressure and pulse data are collected legally and compliantly via the YIWETONG AI All-in-one Machine for intelligent medical record establishment, with explicit user authorization. The raw data includes: unique device identifier (health device ID), system-generated anonymous user identifier (health device monitoring user ID), diastolic blood pressure (mmHg), systolic blood pressure (mmHg), pulse (beats per minute), measurement timestamp (accurate to the minute, defined as "blood pressure collection time"), and system-autogenerated processing exception status flag.
2. Data Processing:
1) Clean the raw data by taking "health device ID" and "blood pressure collection time" as the composite primary key, eliminating entries with format errors or values significantly outside the physiological range (e.g., systolic blood pressure <70 mmHg or >250 mmHg, diastolic blood pressure <40 mmHg or >150 mmHg);
2) Calculate multi-dimensional derived indicators:
- Individual-level indicators: daily average systolic/diastolic blood pressure, blood pressure control compliance rate (target <140/90 mmHg), frequency of hypertensive events, coefficient of variation of pulse;
- Device-level indicators: daily average number of served users, device online rate, frequency of exception reports;
- Population-level indicators: regional hypertensive detection rate, distribution of people with high pulse;
3) Generate a structured and standardized blood pressure measurement dataset that supports trend analysis, risk early warning and model training.
3. Data Application: Output high-quality multi-user blood pressure time-series data, which can be used to generate personal blood pressure trend reports, trigger automatic hypertension early warnings, assist doctors in adjusting treatment plans, evaluate the effectiveness of community hypertension management, and train AI models for cardiovascular disease risk prediction.

搜集汇总

背景与挑战

背景概述

血压测量数据集是一个16KB规模、月度更新的卫生领域数据集，来源于多场景设备采集，用于记录和分析血压与脉搏数据，支持高血压筛查、慢病管理及AI模型训练。数据集通过匿名化处理保障隐私，并经过清洗和衍生指标计算，生成结构化数据以应用于社区健康监护和公共卫生干预。