基于贝叶斯网络的体域网多模态健康数据融合方法

体域网作为无线传感器网络在生物医学领域的一个重要分支能够远程实时监测人体多项健康数据.针对基于体域网采集到的多模态健康数据融合与分析方法进行研究,设计了一套包括动态心电传感器、血压传感器和血氧饱和度传感器的体域网组网方式,提出了一种基于贝叶斯网络模型和推理算法的心肌缺血监测识别方法.通过对60例确诊心脏病患者施行单一模态动态心电监测和多模态健康数据监测对比实验,验证了所提出的多模态健康数据融合方法能够有效提高无症状性心肌缺血的检出率,为临床应用提供了一种新的辅助判别手段.

A Multi-Modal Health Data Fusion Method in Body Sensor Networks Based on Bayesian Networks

As an important branch of wireless sensor networks(WSNs)in biomedical field,body sensor networks ( BSNs) could remotely monitor a variety of human health data in real time. In this paper,we study a multi-modal health data fusion method based on the data collected in BSNs,in which we design a networking for BSNs including Holter sensor,blood pressure sensor and oxygen saturation sensor,and propose a method of myocardial ischemia mo-nitoring and identification based on Bayesian network model and reasoning algorithm. Single-modal Holter monitoring and multi-modal health monitoring were performed in 60 patients with confirmed heart disease,and it was proved that the proposed multi-modal health data fusion method could effectively improve the detection rate of a-symptomatic myocardial ischemia,providing a new auxiliary judgment method for clinical application.