CAREER: Computational Optics and Photonics for Deep Imaging of Live Tissue

职业：用于活组织深度成像的计算光学和光子学

• 批准号: 1750970
• 负责人: Heidy Sierra
• 金额: $ 49.89万
• 依托单位: University of Puerto Rico Mayaguez
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2018
• 资助国家: 美国
• 起止时间: 2018-05-01 至 2024-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1750970&HistoricalAwards=false
• 关键词: CAREER; Computational; Optics; Photonics; Deep

Innovation in non-invasive imaging techniques is part of the effort to provide rapid screening, diagnosis as well as to guide treatment in numerous settings that aspire to offer affordable and efficient healthcare. Most of the existing high-resolution methods are effective primarily on thin and nearly homogeneous transparent samples or over tissue surface. In most realistic scenarios, it is important to acquire information at depth within tissue. High-resolution volumetric imaging approaches may require expensive computational tools for data analysis and complex hardware configurations. Computational optics grounded on signal processing and image reconstruction concepts offers promising alternatives. This research contributes to advance the related state-of-the-art in translational cyberinfrastructure and biomedical technology. Results from this research can improve non-invasive imaging systems for research and patient care while supporting the NSF mission to promote the progress of science and advance the national health. The development of this project involves multidisciplinary efforts from computer science, bioengineering and electrical engineering as well as educational activities with the participation of students from underrepresented groups. This project focuses on providing a framework to support advances on optical imaging techniques that can perform at the needed resolution and speed for various scenarios such as healthcare and biomedical research. The research plan is geared to creating an advanced cyberinfrastructure with simulation and analysis tools to build a computational optical system for deep imaging of live tissues. The components of the framework include three-dimensional optical imaging models employing nonlinear scattering theory that integrate tissue optical properties to characterize their effect into the imaging resolution performance. Additionally, it includes the integration of light-tissue-interaction modeling parameters with compressive sensing concepts and machine learning algorithms for advanced data management. This project targets realistic challenges in biomedical research, including (i) a gap between complex physics of light propagation in tissues and the design of efficient high-resolution imaging systems, (ii) computational optics and photonics for deep imaging of live tissues, and (iii) integration with reliable and state-of-the-art data analytics and visualization environments. The simulations and computational optics tools focus on confocal imaging of skin tissue, which is widely used in biomedical research, and is potentially adoptable in the clinic to guide diagnosis of skin conditions. The education plan addresses three major areas: i) research training and experiences for graduate and undergraduate students, i) course development in topics related with computational optics and data analytics, and iii) outreach to K-12 students and professionals to introduce research issues and opportunities in computational imaging and data analytics.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.

非侵入性成像技术的创新是提供快速筛查、诊断以及指导治疗的努力的一部分，这些环境渴望提供负担得起的高效医疗保健。现有的大多数高分辨率方法主要是对薄的和接近均匀的透明样品或组织表面有效的。在大多数现实场景中，获取组织内部深度信息是很重要的。高分辨率的体积成像方法可能需要昂贵的计算工具来进行数据分析和复杂的硬件配置。基于信号处理和图像重建概念的计算光学提供了很有前途的替代方案。这项研究有助于推进翻译网络基础设施和生物医学技术的相关最新水平。这项研究的结果可以改进用于研究和患者护理的非侵入性成像系统，同时支持NSF促进科学进步和促进国民健康的使命。该项目的发展涉及计算机科学、生物工程和电气工程的多学科努力，以及有来自代表性不足群体的学生参与的教育活动。该项目的重点是提供一个框架，以支持光学成像技术的进步，这些技术可以在各种情况下以所需的分辨率和速度执行，例如医疗保健和生物医学研究。该研究计划旨在创建一个具有模拟和分析工具的先进网络基础设施，以建立一个用于活体组织深度成像的计算光学系统。该框架的组成部分包括使用非线性散射理论的三维光学成像模型，该模型将组织光学属性整合到成像分辨率性能中来表征它们的影响。此外，它还包括光-组织相互作用建模参数与压缩传感概念的集成，以及用于高级数据管理的机器学习算法。该项目针对生物医学研究中的现实挑战，包括(I)组织中光传播的复杂物理与高效高分辨率成像系统设计之间的差距，(Ii)用于活体组织深度成像的计算光学和光子学，以及(Iii)与可靠和最先进的数据分析和可视化环境的集成。模拟和计算光学工具侧重于皮肤组织的共焦成像，这在生物医学研究中得到了广泛的应用，并有可能在临床上用于指导皮肤状况的诊断。教育计划涉及三个主要领域：i)研究生和本科生的研究培训和经验，i)计算光学和数据分析相关主题的课程开发，以及iii)面向K-12学生和专业人员介绍计算成像和数据分析方面的研究问题和机会。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。

项目成果

A Framework For An Artificial Neural Network Enabled Single Pixel Hyperspectral Imager

• DOI: 10.1109/whispers.2019.8921054
• 发表时间: 2019-09
• 期刊: 2019 10th Workshop on Hyperspectral Imaging and Signal Processing: Evolution in Remote Sensing (WHISPERS)
• 作者: Fernando X. Arias;H. Sierra;Emmanuel Arzuaga

Residual Neural Network Architectures to Improve Prediction Accuracy of Properties of Materials

残差神经网络架构可提高材料特性的预测精度

• DOI: 10.1109/bigdata50022.2020.9377934
• 发表时间: 2020
• 期刊: IEEE International Conference on Big Data
• 作者: Tatis D.;Sierra H;Arzuaga H.
• 通讯作者: Arzuaga H.

RCMDD: A Denoising Architecture for Improved Recovery of Reflectance Confocal Microscopy Images of Skin from Compressive Samples

RCMDD：一种降噪架构，可改善压缩样本中皮肤反射共焦显微图像的恢复

• DOI: 10.1109/healthcom46333.2019.9009434
• 发表时间: 2020
• 期刊: Application & Services (HealthCom
• 作者: Arias, Fernando X.;Sierra, Heidy;Arzuaga, Emmanuel
• 通讯作者: Arzuaga, Emmanuel

A Pixel Level Scaled Fusion Model to Provide High Spatial-Spectral Resolution for Satellite Images Using LSTM Networks

• DOI: 10.1109/whispers.2019.8921269
• 发表时间: 2019-09
• 期刊: 2019 10th Workshop on Hyperspectral Imaging and Signal Processing: Evolution in Remote Sensing (WHISPERS)
• 作者: Carlos A. Theran;Michael A. Álvarez;Emmanuel Arzuaga;H. Sierra

Studying the effect of a decision fusion model on enhanced hyperspectral images

• DOI: 10.1117/12.2560107
• 发表时间: 2020-05
• 作者: Carlos A. Theran;Michael A. Álvarez;H. Sierra;Emmanuel Arzuaga