VARIABLE VOXEL-SIZE INTENSITY CORRECTION

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。 高密度表面线圈阵列已被用于获得用于大脑、乳房 和心脏MRI研究[1，2]。由于线圈元件的布局和大小，这些阵列通常表现出灵敏度 整个视场的变化。在文献中已经研究了几种强度校正方案，其中许多 这些算法在处理过程中显著改变了空间噪声分布，这是不希望的[3，4]。在这项研究中， 我们提出了一种变体素大小强度校正方法，它在保持原始噪声轮廓的同时降低了 整个图像的强度变化。 基本思想包括平均低敏感度区域中的体素，并适当缩放结果，从而提高 这些地区的SNR。我们给出了这种变体素大小算法的结果，用于超高分辨率的3D 乳房患者使用16通道表面线圈阵列进行扫描。我们假设线圈的强度分布是一个缓慢变化的 空间函数，并在原始数据上使用低通滤波来近似该偏置场。我们将这个偏置场归一化 并创建了体素体积图，其中较大的体素大小对应于偏置场的低强度区域。我们 通过不同带宽的缩放低通滤波器分别传递原始数据集，以实现所需的 体素体积贴图中的分辨率。在最终校正的数据集中，我们为每个基于体素的数据选择了合适的数据集 在体素体积贴图上。 要了解卢卡斯中心正在进行的其他项目，请访问http://rsl.stanford.edu/(卢卡斯年度报告和 ISMRM 2011年摘要)

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. High-density surface coil arrays have been used for attaining highly-accelerated, high resolution images for brain, breast and cardiac MRI studies [1,2]. Due to the layout and size of the coil elements, these arrays usually exhibit sensitivity variations across the field of view. Several intensity correction schemes have been investigated in literature, and many of these algorithms significantly change the spatial noise profile during the process, which is undesirable [3,4]. In this study, we present a variable voxel-size intensity correction method that maintains the original noise profile while reducing intensity variations across the image. The principle idea consists of averaging voxels in regions of low sensitivity and properly scaling the result, thereby boosting the SNR in these regions. We present the results of this variable voxel-size algorithm for an ultrahigh resolution, 3D breast patient scan using a 16-channel surface coil array. We assumed that the coil intensity profile is a slowly varying spatial function and approximated this bias field using a low pass filter on the original data. We normalized this bias field and created a "voxel volume" map where larger voxel sizes correspond to regions of low intensity of the bias field. We passed the original dataset individually through scaled low pass filters of varying bandwidths to achieve the desired resolution in the voxel volume map. In the final corrected dataset, we chose the appropriate dataset for each voxel based on the voxel volume map. To read about other projects ongoing at the Lucas Center, please visit http://rsl.stanford.edu/ (Lucas Annual Report and ISMRM 2011 Abstracts)