CAREER: Blind-Label and Label-Free Acoustic Far-Field Subwavelength Imaging

这项学院早期职业发展(Career)补助金将启用新的成像方法，将硬件和软件创新结合在一起，显著提高声成像的分辨率和实用性。在生物医学诊断中，本研究所寻求的成像系统将提高医学超声成像质量，扩大检查应用。人体和动物体内的位置，如深部脑组织、亚毫米大小的肾结石和血管壁上的病变，使用现有的超声波技术很难获得，将通过新开发的成像系统进行访问。在水下传感方面，这项研究将为水下成像/通信、水下机器人/航行器、水下基础设施维护、渔业、海洋生物和水下自然资源勘探提供增强的传感能力。在无损检测方面，这项研究将提高农业和采矿业的土壤和岩石成像质量，并为制造业的质量控制提供支持。该项目的教育目标是促进和扩大对跨学科声学研究和各级STEM教育的参与。这一目标将通过以下方式实现：开发高中跨学科声学教育工作坊，为本科生，特别是来自代表不足的群体的本科生，创建一个将本科课程与实验室项目相结合的渠道，并通过互动展览和公开演示向公众展示跨学科声学研究。该项目将为解决远场声学成像中长期存在的分辨率和穿透深度之间的权衡提供新的解决方案。大多数现有的亚波长成像方法解决了这种权衡，需要外部的“标签”，如超材料和造影剂，放置在物体附近。这些标签需要是静态的，或者在成像期间被精确跟踪。因此，目前的标记沉积在实际应用中受到极大的限制。该项目将提供盲标签和无标签解决方案，以克服这些限制。将在外部标记沉积可行的情况下开发盲标记法，例如在浅水中成像以及血管内和血管周围区域的成像。无标记方法将被开发用于外部标记沉积困难和不切实际的情况，例如深海声纳成像、无损结构检查和深层组织区域的生物医学成像。盲标签和无标签方法都将遵循硬件和软件协同设计框架，该框架由以下两部分组成：1)空间混合：基于随机沉积的声散射体或无标签环境不均匀的盲标签的物理成像过程，以及2)计算重建：使用与空间波混合过程共同优化的算法进行计算信号处理和成像重建。除了概念和方法的发展，还将开发模拟特定应用环境和硬件/软件条件的实验室成像系统，以验证建议的生物医学超声、水下声纳成像和结构无损检测领域的盲标和无标号方法。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

This Faculty Early Career Development (CAREER) grant will enable new imaging methodologies that combine both hardware and software innovations to significantly improve the resolution and the practicality of acoustic imaging. In biomedical diagnosis, the imaging systems sought in this research will improve medical ultrasound imaging quality and expand examination applications. Locations in human and animal bodies, such as deep brain tissues, kidney stones with sub-millimeter sizes, and lesions on the blood vessel walls, which are difficult to access using the existing ultrasound technologies, will be accessible via the newly developed imaging systems. In underwater sensing, this research will provide enhanced sensing capabilities for underwater imaging/communication, underwater robots/vehicles, underwater infrastructure maintenance, fishery, marine biology, and underwater natural resource exploration. In non-destructive inspection, this research will enhance soil and rock imaging quality in agriculture and the mining industry, and support quality control in the manufacturing industry. The educational objective of this project is to promote and broaden participation in interdisciplinary acoustics research and STEM education at all levels. This objective will be accomplished through developing an education workshop for high school interdisciplinary acoustics education, creating a pipeline that combines undergraduate courses with lab-based projects for undergraduate students, especially those from underrepresented groups, and demonstrating interdisciplinary acoustics research to general public through interactive exhibitions and public demonstrations.This project will provide new solutions to the long-existing tradeoff between resolution and penetration depth in far-field acoustic imaging. Most of the existing subwavelength imaging approaches that address this trade-off require exogenous “labels”, such as metamaterials and contrast agents, to be deposited close to the objects. Those labels need to be either static or precisely tracked during imaging. Therefore, current label depositions are extremely restrictive in practical applications. This project will provide blind-label and label-free solutions to overcome the restrictions. Blind-label approaches will be developed for situations where the external label deposition is viable, e.g., imaging in shallow water and imaging of areas in and around blood vessels. Label-free approaches will be developed for situations where the external label deposition is difficult and impractical, such as deep underwater sonar imaging, non-destructive structure inspection, and biomedical imaging of deep tissue regions. Both blind-label and label-free approaches will follow a hardware and software co-design framework that consists of the two parts as follows: 1) spatial mixing: a physical imaging process based on blind labels, which are randomly deposited acoustic scatterers, or label-free environmental inhomogeneities, and 2) computational reconstruction: computational signal processing and imaging reconstruction using algorithms co-optimized with the spatial wave mixing process. Beyond concept and methodology developments, in-lab imaging systems emulating application-specific environmental and hardware/software conditions will be developed to validate the proposed blind-label and label-free approaches in the areas of biomedical ultrasonography, underwater sonar imaging, and structure non-destructive testing.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.