SGER: Fast Iterative Methods for Diffuse Optical Tomography

SGER：漫反射光学断层扫描的快速迭代方法

• 批准号: 0353160
• 负责人: Birsen Yazici
• 金额: $ 5.34万
• 依托单位: Rensselaer Polytechnic Institute
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2003
• 资助国家: 美国
• 起止时间: 2003-08-15 至 2005-06-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0353160&HistoricalAwards=false
• 关键词: SGER; Fast; Iterative; Methods; Diffuse

0332892YaziciThe goal of Diffuse Optical Tomography (DOT) is to reconstruct spatial maps of the optical properties of tissue from the measurements of transmitted light based on a forward model of the photon propagation. Due to diffuse nature of light photons, DOT image reconstruction is a challenging 3D problem with a relatively large number of unknowns and limited measurements. The general objective of the proposed project is to develop high resolution and fast DOT reconstruction algorithms by developing an optimization based iterative reconstruction scheme that is spatially adaptive given high resolution a priori anatomical information. The specific objectives of this small exploratory grant are to 1) investigate adaptive multigrid/multiscale and domain decomposition methods for the diffuse optical image reconstruction problem and 2) evaluate, benchmark and optimize domain decomposition and adaptive multigrid image reconstruction algorithms for optical absorption tomography with simulated, experimental and in vivo clinical breast data collected by a novel hybrid MRI and time resolved optical imager. The proposed method is expected to provide more accurate and high resolution reconstructions in less time, which would have a major impact on all aspects of clinical breast cancer management and has the potential to catalyze innovative advances in other medical imaging and multimodality imaging techniques.

漫射光学层析成像（DOT）的目标是基于光子传播的正演模型，从透射光的测量中重建组织光学特性的空间图。由于光子的漫射特性，DOT图像重建是一个具有相对大量未知和有限测量的具有挑战性的3D问题。该项目的总体目标是通过开发一种基于优化的迭代重建方案来开发高分辨率和快速的DOT重建算法，该方案在给定高分辨率先验解剖信息的情况下具有空间适应性。这项小型探索性资助的具体目标是：1)研究用于漫反射光学图像重建问题的自适应多网格/多尺度和区域分解方法；2)评估、基准和优化用于光学吸收断层扫描的区域分解和自适应多网格图像重建算法，这些算法使用由新型混合MRI和时间分辨光学成像仪收集的模拟、实验和体内临床乳房数据。所提出的方法有望在更短的时间内提供更准确和高分辨率的重建，这将对临床乳腺癌管理的各个方面产生重大影响，并有可能催化其他医学成像和多模态成像技术的创新进展。