Remotely Powered Flexible Wearable Sensors for Managing Chronic Diseases

据世界卫生组织称，心血管疾病是全球头号死亡原因。正如加拿大心脏和中风基金会2019年的报告所指出的那样，加拿大每五分钟就有一人死于心脏病、中风或血管性认知障碍。这一死亡率超过了其他疾病，即死于心脏病、中风和血管性认知障碍的人数比死于所有癌症的人数总和多13%。2007年至2017年期间，加拿大有260万人因心血管疾病住院。仅心血管疾病一项，每年的直接（医疗）和间接（收入损失）成本总计达212亿美元。中风每年给加拿大经济造成36亿美元的损失，包括医生服务、医院费用、工资损失和生产力下降。痴呆症（血管性认知障碍的最严重形式）每年的直接和间接成本为330亿美元，预计到2040年这一数字将攀升至每年2930亿美元。随着信息技术的进步，借助可穿戴无线生物医学传感器，可以在早期预防和发现心血管疾病。在过去的十年中，许多类型的可穿戴传感器被开发出来。但是，它们都没有提供可与医院医疗级设备记录的数据相媲美的信息。这主要是由于人们在使用可穿戴设备进行日常活动时，由于运动伪影导致的信号失真。运动伪影是由以下几个因素引起的：(1)传感器的重量会使附着在皮肤上的电极产生信号失真；(2)由于运动和肌肉活动导致皮肤电极接触电阻的变化。该项目旨在创造一种创新的解决方案，解决现有可穿戴传感器面临的运动伪影问题。我们建议在以下方面进行研究：(1)符合人体形状的柔性薄电极，改善接触阻抗；(2)适用于柔性电极中使用的干电极的超高输入阻抗模拟前端接口电路。这将有助于最大限度地提高输入信号强度，同时最大限度地减少环境噪声；(3)容忍大运动伪影的大动态范围仪表放大器，即在输入端出现大运动伪影时避免放大器饱和；(4)取下传感器内的电池，通过远程给传感器供电来减轻重量；(5)传感器超低功耗嵌入式机器学习算法，用于补偿运动伪影和CVD预警信号检测。目标是为远程供电的柔性可穿戴传感器设计一种系统级芯片解决方案，该解决方案薄、亲肤、可弯曲、重量轻、成本低。该项目的成功将为慢性病的预防、预测和管理创造一个解决方案。这将有助于提高心血管疾病患者的生活质量，同时有助于降低心血管疾病相关的医疗费用。

Cardiovascular diseases (CVDs) are the number ONE cause of death globally according to World Health Organization. As pointed out in the 2019 Report from Heart & Stroke Foundation of Canada, one person dies in Canada every five minutes from heart conditions, stroke or vascular cognitive impairment. This death rate outpaces other diseases, i.e. 13% more people die of heart condition, stroke and vascular cognitive impairment than die from all cancers combined. 2.6 million CVD related hospitalizations in Canada between 2007 and 2017. CVDs alone cost totaling $21.2 billion a year in direct (medical) and indirect (lost earnings) costs. Stroke costs the Canadian economy $3.6 billion a year in physician services, hospital costs, lost wages and decreased productivity. Dementia (the most severe form of vascular cognitive impairment) costs $33 billion a year in direct and indirect costs and this number is expected climb to $293 billion a year by 2040. With the advancement of information technologies, it is possible to prevent and detect CVDs at early stage with the help of wearable wireless biomedical sensors. Many types of wearable sensors were developed in the last decade for this purpose. However, none of them provides the information that are comparable to data recorded by medical grade equipment at hospital. This is mainly due to signal distortions caused by motion artifacts when people carry out daily activities with wearable devices. The motion artifacts are caused by several factors: (1) the weight of sensor that pulls the electrodes attached to skin causing signal distortion, and (2) the change of skin-electrode contact resistance due to motion and muscle activities. This project aims to create an innovative solution that solves the motion artifact problems faced by existing wearable sensors. We propose to conduct research in following areas: (1) thin flexible electrodes that conform to body shape with improved contact impedance; (2) ultra-high input impedance analog front-end interface circuit that is suitable for dry electrodes used in flexible electrodes. This will help to maximize input signal strength while minimizes the ambient noises; (3) large dynamic range instrumentation amplifier that is tolerant to large motion artifacts, i.e. to avoid saturation of amplifier when a large motion artifact is presented at input; (4) removal of battery in the sensor to reduce the weight by powering the sensor remotely; (5) ultra-low power embedded machine learning algorithm at the sensor for compensating motion artifacts and for CVD early warning sign detection. The goal is to design a system-on-chip solution for a remotely powered flexible wearable sensors that is thin, skin-friendly, bendable, light weight, and low cost. The success of proposed project will create a solution for preventing, predicting, and managing chronic diseases. It will help to improve the quality of CVD patient's life while helping reduce the CVD related health care cost.