Enabling Population-Scale Physical Activity Measurement on Common Mobile Phones

描述(由申请人提供)： 这个U01项目的主要目标是硬件和软件技术的技术开发、部署和评估，这些技术将使人们能够使用普通手机对身体活动进行人口规模的纵向测量。在亚洲上市的手机已经具备内置加速计和低功率无线通信功能，不久还会在美国上市。这项研究将调查如何使用这些计算设备来准确测量身体活动的类型、强度和比赛持续时间。通过利用消费者在许多美国人将购买、维护和携带的手机上的支出，有可能进行大规模研究，以负担得起的成本测量数十万参与者的体力活动并对其进行数月或数年的远程监控。麻省理工学院设计的无线加速计将被重新设计，这样它们就可以向2011年上市的普通手机发送数据。使用当前版本传感器的实验室测试将用于比较通过将电话加速度计数据与通过在不同方便的身体位置(例如，在手表或包中、在鞋子上、在臀部等)佩戴一个或多个无线传感器获得的数据相结合而获得的相对信息增益。将确定加速计的最佳但实用的配置，使得在移动电话上运行的软件可以自动检测特定的身体活动，例如快步行走、跑步、骑自行车、爬楼梯、扫地、进行运动等。传感器和软件设计将解决的技术挑战包括：(1)获得实际电池寿命，(2)获取高时间分辨率的身体活动数据，(3)使检测算法能够针对个人定制，(4)解决实际终端用户对人体工程学、舒适性和社会可接受性的担忧，(5)允许对具有数十万电话用户的研究进行实时和低成本的远程监测和维护，以及(4)允许使用其他现成的传感器设备，如心率监测器，当它们可用时。将采用参与式设计过程来制定战略，以获得典型电话用户的纵向遵守。经过两轮迭代的技术开发，每一轮都在斯坦福大学进行实验室验证，该技术将在50名典型电话用户中部署10个月。将测量相对于自我报告、可接受性和纵向遵从性的有效性。

DESCRIPTION (provided by applicant): The primary aim of this U01 project is the technical development, deployment, and evaluation of hardware and software technology that will enable population-scale, longitudinal measurement of physical activity using common mobile phones. Mobile phones available in Asia and soon in the U.S. already include internal accelerometers and low-power wireless communication capabilities. This study will investigate how to use these computing devices for accurate measurement of physical activity type, intensity and bout duration. By exploiting consumer expenditures on phones that many Americans will purchase, maintain, and carry, it may be possible to run large-scale studies where the physical activity of hundreds of thousands of participants is measured and remotely monitored for months or years at an affordable cost. Wireless accelerometers designed at MIT will be redesigned so that they can send data to common mobile phones available in 2011. Laboratory testing using the current version of the sensors will be used to compare the relative information gain that can be obtained by combining the phone accelerometer data and data obtained by wearing one or more wireless sensors on different convenient body locations (e.g., in a watch or bag, on a shoe, at the hip, etc.). Optimal but practical configurations of accelerometers will be determined so that software running on the mobile phone can automatically detect specific physical activities such as brisk walking, running, cycling, climbing stairs, sweeping, playing sports, etc. Technical challenges that will be addressed by the sensor and software design include, (1) obtaining practical battery life, (2) acquiring physical activity data at high temporal resolution, (3) enabling person-specific customization of the detection algorithms, (4) addressing practical end-user concerns about ergonomics, comfort, and social acceptability, (5) permitting real-time and low-cost remote monitoring and maintenance for studies with hundreds of thousands of phone users, and (4) enabling use of other off-the-shelf sensor devices, such as heart rate monitors, as they become available. A participatory design process will be employed to develop strategies for obtaining longitudinal compliance from typical phone users. After two rounds of iterative technical development, each with laboratory validation conducted at Stanford, the technology will be deployed with 50 typical phone users for 10 months. Validity relative to self report, acceptability, and longitudinal compliance will be measured.