CSR: Small: Transforming Mobile Devices into Active Sonar Systems for Medical Applications

该项目利用现有移动设备上常见的扬声器和麦克风，将包括智能手机在内的移动系统转换为主动声纳系统。使用这种方法，该系统解决了两个关键医疗保健领域长期存在且日益紧迫的问题：药理学诱导的呼吸暂停诊断和肌肉疾病患者的辅助接口。与在兆赫频率下工作的医疗超声设备不同，智能手机上的麦克风和扬声器的设计频率要低得多。该项目回答了在智能手机上使用主动声纳解决医疗应用可行性的基本问题。它将产生跟踪呼吸信号的算法，这是药理学诱导的呼吸暂停检测所必需的，即使在身体其他部位有运动的情况下也是如此。最后，该项目将产生一种算法，利用智能手机处理的声音信号的反射，从口腔外跟踪舌头的运动。导致致命呼吸暂停事件的医疗条件日益成为突发公共卫生事件。该项目如果成功，将实现一个能够检测这些事件的移动系统，从而提供一个可能挽救无数生命的机会。该项目的第二部分将创建一个基于舌头的交互工具包，该工具包将帮助成千上万的肌肉残疾患者仅使用一个软件应用程序就能与现有的移动设备进行交互——目前还不存在这种功能。最终的原型将在DawgBytes夏令营和华盛顿大学妇女倡议中展示。项目的结果将被添加到项目网页。数据将被剥离任何私人信息，并将被添加到项目网页。由此产生的呼吸信号和我们部署的基准数据将被添加到网页中进行独立分析。该项目网页如下：http://apnea.cs.washington.eduThis该奖项反映了美国国家科学基金会的法定使命，并通过基金会的智力价值和更广泛的影响审查标准进行评估，认为值得支持。

The project converts mobile systems including smartphones into active sonar systems by leveraging the speakers and microphones commonly available on existing mobile devices. Using this approach, the system addresses long-standing and increasingly urgent problems in two key healthcare domains: pharmacologically-induced apnea diagnosis and assistive interfaces for patients with muscular disorders. Unlike medical ultrasound devices that operate in the megahertz frequencies, microphones and speakers on smartphones are designed for much lower frequencies. The project answers fundamental questions about the feasibility of using active sonar on smartphones to address medical applications. It will produce algorithms that track breathing signals, required for pharmacologically-induced apnea detection, even in the presence of movements from other parts of the body. Finally, the project will result in algorithms that achieve tracking of the tongue motion from outside the mouth using the reflections from the acoustic signals as processed by the smartphone. Medical conditions that result in fatal apnea events have increasingly become a public health emergency. The project, if successful, would achieve a mobile system that can detect these events, thus providing an opportunity to potentially saving numerous lives. The second part of the project will create a tongue-based interaction toolkit that would help hundreds of thousands of patients with muscular disabilities to interact with existing mobile devices, using only a software app - a capability that currently does not exist. The resulting prototypes will be showcased at DawgBytes Summer Camps and University of Washington Women's Initiative. The results of the project will be added to the project webpage. The data will be stripped of any private information and will be added to the project webpage. The resulting breathing signals and the benchmarks data from our deployments will be added to the webpage for independent analysis. The project webpage is at the following link: http://apnea.cs.washington.eduThis award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.