BAYESIAN RECONSTRUCTION FROM MULTICHANNEL K-SPACE DATA USING GRAPH-CUT ALGORITHM

这个子项目是利用资源的许多研究子项目之一。 由NIH/NCRR资助的中心拨款提供。对子项目的主要支持 子项目的首席调查员可能是由其他来源提供的， 包括美国国立卫生研究院的其他来源。为子项目列出的总成本可能 表示该子项目使用的中心基础设施的估计数量， 不是由NCRR赠款提供给次级项目或次级项目工作人员的直接资金。 目的1.一种用于脑图像分割的图割算法 该项目旨在获得用于结构脑图像分割的新算法和软件。我们将同时考虑单通道数据和多通道数据。该项目将把图像分割成3个组织类别：白质、灰质和脑脊液。我们将使用图割方法，这是计算机视觉中流行的分割任务。 目的2.基于图割的并行核磁共振贝叶斯重建 该项目旨在利用计算机科学和图论的最新进展，为重建磁共振并行成像(MRPI)数据开发一种新的算法范式。具体地说，我们将进一步开发和改进计算效率高的mri贝叶斯方法，这些方法有可能克服传统mr成像的基本限制。 目的3.一种用于脑图像分割的图割算法 重建核心中的这个子项目旨在改进图像分割。具体地说，目标是开发基于图割的新算法，用于完全联合图像的分割、配准和偏场校正

This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. Primary support for the subproject and the subproject's principal investigator may have been provided by other sources, including other NIH sources. The Total Cost listed for the subproject likely represents the estimated amount of Center infrastructure utilized by the subproject, not direct funding provided by the NCRR grant to the subproject or subproject staff. Aim 1. A Graph Cut Algorithm For Brain Image Segmentation This project aims to obtain novel algorithms and software for structural brain image segmentation. We will consider both single-modality as well as multi-modality data. The project will result in image segmentation into 3 tissue classes: white matter, gray matter and cerebro-spinal fluid. We will use a graph cut approach, which is popular is computer vision for segmentation tasks. Aim 2. Bayesian Reconstruction of Parallel MRI Using Graph Cuts This project aims to develop a new algorithmic paradigm for the reconstruction of MR Parallel Imaging (MRPI) data by using recent advances in Computer Science and Graph Theory. Specifically, we will further develop and refine computationally efficient Bayesian methods for MRI that have the potential to overcome fundamental limits of traditional MR imaging Aim 3. A Graph Cut Algorithm For Brain Image Segmentation This subproject in the reconstruction core aims to improve image segmentation. Specifically, the aim is to develop new algorithms based on Graph cuts for completely joint image segmentation, registration and bias field correction