计算机应用 ›› 2019, Vol. 39 ›› Issue (5): 1551-1556.DOI: 10.11772/j.issn.1001-9081.2018110008

• 应用前沿、交叉与综合 • 上一篇    

基于多因素线索长短期记忆模型的血压分析预测

刘晶1,2, 吴英飞1,2, 袁贞明1,2, 孙晓燕1,2   

  1. 1. 杭州师范大学 杭州国际服务工程学院, 杭州 310000;
    2. 移动健康管理系统教育部工程研究中心, 杭州 310000
  • 收稿日期:2018-12-04 修回日期:2019-01-09 出版日期:2019-05-10 发布日期:2019-05-14
  • 通讯作者: 吴英飞
  • 作者简介:刘晶(1993-),女,山西运城人,硕士研究生,主要研究方向:数字医疗、健康管理工程、机器学习、数据挖掘;吴英飞(1972-),男,浙江杭州人,讲师,博士,CCF会员,主要研究方向:图形图像处理、多媒体、人工智能;袁贞明(1972-),男,浙江杭州人,教授,博士,CCF会员,主要研究方向:人工智能、多媒体分析、医学信息学;孙晓燕(1980-),女,浙江杭州人,讲师,博士,主要研究方向:计算机辅助医疗。
  • 基金资助:
    浙江省自然科学基金资助项目(LQ16H180004);杭州市软科学计划项目(20140834M49)。

Blood pressure prediction with multi-factor cue long short-term memory model

LIU Jing1,2, WU Yingfei1,2, YUAN Zhenming1,2, SUN Xiaoyan1,2   

  1. 1. Hangzhou Institute of Service Engineering, Hangzhou Normal University, Hangzhou Zhejiang 310000, China;
    2. Engineering Research Center of Mobile Health Management System, Ministry of Education, Hangzhou Zhejiang 310000, China
  • Received:2018-12-04 Revised:2019-01-09 Online:2019-05-10 Published:2019-05-14
  • Supported by:
    This work is partially supported by the Natural Science Foundation of Zhejiang Province (LQ16H180004), the Project of Hangzhou Soft Science Research Plan (20140834M49).

摘要: 高血压是危害健康的重要因素,为了预防血压突然升高造成严重后果,在传统长短期记忆(LSTM)网络基础上,提出一种多因素线索LSTM模型,适用于血压的短期预测和长期预测,能够对血压的不良变化提前作出预警。模型中用到的多因素线索包括时序数据线索和上下文信息线索(包括个人基本信息和环境信息)两大类,使得血压预测不仅提取血压数据本身的特征,还提取与血压相关联的时序数据变化特征和其他关联属性的数据特征。模型首次将环境因素加入血压预测,并采用多任务学习方式,能够更好地捕捉数据之间隐藏的关联性,提高模型泛化能力。实验结果表明,所提模型相较于传统LSTM模型和添加了上下文信息层的LSTM(LSTM-CL)模型在舒张压的预测误差与预测偏差方面分别降低2.5%,3.8%和1.9%,3.2%,在收缩压的预测误差和预测偏差分别降低0.2%,0.1%和0.6%,0.3%。

关键词: 高血压, 血压预测, 长短期记忆, 时序数据, 上下文信息

Abstract: Hypertension is an important hazard to health. Blood pressure prediction is of great importance to avoid grave consequences caused by sudden increase of blood pressure. Based on traditional Long Short-Term Memory (LSTM) network, a multi-factor cue LSTM model for both short-term prediction (predicting blood pressure for the next day) and long-term prediction (predicting blood pressure for the next several days) was proposed to provide early warning of undesirable change of blood pressure. Multi-factor cues used in blood pressure prediction model included time series data cues (e.g. heart rate) and contextual information cues (e.g. age, BMI (Body Mass Index), gender, temperature).The change characteristics of time series data and data features of other associated attributes were extracted in the blood pressure prediction. Environment factor was firstly considered in blood pressure prediction and multi-task learning method was used to help the model to capture the relation between data and improve the generalization ability of the model. The experimental results show that compared with traditional LSTM model and the LSTM with Contextual Layer (LSTM-CL) model, the proposed model decreases prediction error and prediction bias by 2.5%, 3.8% and 1.9%, 3.2% respectively for diastolic blood pressure, and reduces prediction error and prediction bias by 0.2%, 0.1% and 0.6%, 0.3% respectively for systolic blood pressure.

Key words: hypertension, Blood Pressure (BP) prediction, Long Short-Term Memory (LSTM), time series data, contextual information

中图分类号: