RF infrasonics for internal tissue characteristics

内脏器官、血管、气道和喉部组织力学特性的变化可作为许多引起肿胀、炎症、肿瘤和鼠疫沉积的疾病的诊断和预后指标。除了由生命体征和声音引起的自发振动外，还可以通过外部驱动触发内部组织的机械运动，随后的振动响应和阻尼特性可以在体外观察到。目前基于听诊或听诊器的设备可以检索到高于20hz的组织振动，便携式超声可以检索到低于2hz的特征频率的运动细节，帧率为20hz。然而，许多重要的内部组织振动特性在2 - 20赫兹之间的次声波段是难以准确测量的现有工具，特别是在初级保健和家庭。然而，次声波段涵盖了主要血管和呼吸组织的典型振动特征。例如，当肺部发生炎症时，例如COVID-19肺炎，肺和支气管组织会含有更多的水分，这种情况称为水肿，并且它们的振动和阻尼特性会发生很大变化。当鼠疫在主动脉和冠状动脉内沉积时，振动特征也会发生变化。多年来，超声波和听诊器已经得到了完善，但超越目前的能力来缩小次声范围的差距将是具有挑战性的。该项目将开发所需的硬件和软件，以实现新的功能，并在具有人造组织的人体模型上进行测试。所开发的传感器工具可用于远程医疗或与其他常规诊断和图像工具一起使用，以提高成像质量。研究成果将被翻译成有关传感器的本科课程，并将被制作成面向K-12学生和公众的宣传材料。本研究的主要目标是开发一种可穿戴或便携式射频（RF）传感器，该传感器可以覆盖较宽的组织振动带宽，包括2 - 20hz的临界次声带，以及适合检索时域和频谱域瞬态特征的信号处理算法。无线电信号的采样率可以覆盖非常宽的范围，用于检测非常慢到非常快的组织和器官运动。相比之下，快速的组织运动在有限的帧率下很难进行内部成像，而缓慢的运动则不会发出听诊器接收到的声音。由于许多内部组织具有高粘度和各向异性，因此可以将多个射频传感器部署为多输入多输出（MIMO）网络，以提高空间分辨率。该项目将开发类似于探地雷达的近场MIMO射频传感，以及类似于地震学的声学探测。具体而言，该项目将构建一个具有模仿内部组织的人体幻影，并在幻影上安装MIMO射频传感器阵列。具有高时间和光谱分辨率的振动和阻尼特性的信号处理算法将被开发和基准。从组织激发到具有可扩展几何参数的射频收发器输出的物理模型也将被开发。最后，将对健康参与者进行人体研究，为改进传感器设计提供反馈。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

The change of tissue mechanical properties in internal organs, vessels, airway and larynx can serve as diagnostic and prognostic indicators for many disorders that cause swelling, inflammation, tumors and plague deposition. In addition to the self-initiated vibrations induced by vital signs and egophony, mechanical motions of internal tissues can be triggered through an external actuation, when the subsequent vibration responses and damping characteristics can be observed outside the body. Present devices based on auscultation or stethoscope can retrieve tissue vibration higher than 20 Hz, and portable ultrasound can retrieve motion details with characteristic frequencies lower than 2 Hz with a frame rate of 20 Hz. However, many important internal tissue vibration characteristics in the infrasound band between 2 – 20 Hz are difficult to be accurately measured by present tools, especially in primary care and at home. However, the infrasound band covers the typical vibration characteristics of main vascular and respiratory tissues. For example, when the lung has inflammation, such as in the COVID-19 pneumonia, the lung and bronchi tissues will contain more water, a condition called edema, and have a large change in their vibration and damping characteristics. When plague is deposited inside aortic and coronary arteries, the vibration feature also evolves. Ultrasound and stethoscopes have been perfected over the years, but improvement beyond the present capabilities to close the gap of the infrasound range would be challenging. This project will develop the required hardware and software to achieve new capabilities and test on a manikin with artificial tissues. The developed sensor tools can be used for telemedicine or with other conventional diagnostic and image tools to improve imaging quality. The research results will be translated into undergraduate courses on sensors, and into promotional materials for K-12 students and public. The major objective of this research is to develop a wearable or portable radio frequency (RF) sensor that can cover broad tissue vibration bandwidth including the critical infrasound band of 2 – 20 Hz, together with signal processing algorithms appropriate to retrieve the transient features in both temporal and spectral domains. The sampling rate for radio signals can cover a very broad range for detecting very slow to very fast tissue and organ motions. In comparison, fast tissue motion is difficult for internal imaging with limited frame rates, and slow motion does not make a sound for the stethoscope to pick up. As many internal tissues have high viscosity and anisotropy, multiple RF sensors can be deployed as a multiple-input-multiple-output (MIMO) network to improve the spatial resolution. This project will develop near-field MIMO RF sensing similar to the ground-penetrating radar, and the acoustic detection is similar to seismology. Specifically, the project will construct a body phantom with imitation internal tissues a MIMO RF sensor array on the phantom. Signal processing algorithms for vibration and damping characteristics with high temporal and spectral resolutions will be developed and benchmarked. A physical model from the tissue excitation to the RF transceiver outputs with scalable geometrical parameters will also be developed. Lastly, human study with healthy participants will be conducted to provide feedbacks for improvement on sensor design.This 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.