低负荷生理信息动态提取方法及其在睡眠健康评估中应用的研究

本项目深入探讨低负荷生理信息动态提取的理论和方法，并针对睡眠健康评估与预警展开应用研究。项目采用可穿戴微型传感器构成的体域网，实现对心电（ECG）、多通道生物电阻抗等生理信息的低负荷监测和特征提取。在此基础上，建立睡眠健康评估与预警模型，研究睡眠危险因子与模型之间的相互关系。研究内容主要包括：1） 低负荷生理信息动态监测方法；2）高可信度健康指标特征提取算法；3）高稳健性睡眠健康评估与预警模型。关键科学问题是低负荷生理信息动态提取的全局误差最小化问题和睡眠健康评估与预警模型稳健性问题。创新点主要体现在：1）从传感方式、伪差去除、可信度增强算法等多方面系统研究低负荷生理信息动态提取方法及睡眠健康评估方法；2）将低负荷监测方法与睡眠健康评估有机结合。通过本项目研究，一方面为低负荷生理信息监测开辟更广阔的应用空间，另一方面有助于探索睡眠生理机理，进而为全民低成本健康实用技术的开发提供理论基础。

The project researched deeply the theory and method of ubiquitous physiological information dynamic extraction and its application on sleep health evaluation and early-warning. The project realized ubiquitous physiological information monitoring (e.g. ECG, Multi-bioimpedance measurements) and feature extraction by using wearable micro-sensors based on body-area sensor networks (BSN). Then, the model of sleep health evaluation and warning was established, and the relationship between risk factors and the model was deeply researched. The project mainly studies: 1) ubiquitous physiological information dynamic monitoring method; 2) high reliability health indexes dynamic extraction method; 3) high robustness model for sleep health evaluation and early-warning. The key scientific problems: system error minimization and model robustness. Main innovations: 1) physiological information dynamic extraction and sleep health evaluation were systematically investigated by sensing method, removing motion artifact and credibility enhancement algorithm. 2) organically integrated of physiological information monitoring and sleep health evaluation model. The project will expand the field of application for ubiquitous physiological information monitoring. On the other hand, it will contribute to exploring sleep health physiological mechanism and providing theoretical basis for practical technique of low cost health.