Automatic macroscopic density artefact removal in a Nissl-stained microscopic atlas of whole mouse brain
Automatic macroscopic density artefact removal in a Nissl-stained microscopic atlas of whole mouse brain
复制标题
自动去除小鼠全脑尼氏染色显微图谱中的宏观密度伪影
DOI：
10.1111/jmi.12058
复制
发表时间：
2013-08-01
影响因子：
2
通讯作者：
Gong, H.
中科院分区：
文献类型：
--
作者：
Ding, W.;Li, A.;Gong, H.
在微米尺度上获取整个小鼠大脑是一个复杂、连续和耗时的过程。由于由样品制备和显微镜检查引起的缺陷，所获取的图像数据集遭受各种宏观密度伪影，其使图像质量恶化。我们必须开发可用的预处理方法，通过去除影响细胞分割、血管跟踪和可视化的伪影来提高图像质量。在这项研究中，一套自动伪影去除方法，提出了组织染色和光学显微镜获得的图像。这些方法显著改善了包含各种结构（包括细胞和血管）的复杂图像。测试了具有Nissl染色的整个小鼠脑数据集，并且处理后的图像的强度均匀地分布在不同的脑区域中。此外，经过处理的图像数据集具有均匀的亮度和高质量，现在是用于图像分析的基本图谱，包括细胞分割，血管跟踪和可视化。
Acquiring a whole mouse brain at the micrometer scale is a complex, continuous and time-consuming process. Because of defects caused by sample preparation and microscopy, the acquired image data sets suffer from various macroscopic density artefacts that worsen the image quality. We have to develop the available preprocessing methods to improve image quality by removing the artefacts that effect cell segmentation, vascular tracing and visualization. In this study, a set of automatic artefact removal methods is proposed for images obtained by tissue staining and optical microscopy. These methods significantly improve the complicated images that contain various structures, including cells and blood vessels. The whole mouse brain data set with Nissl staining was tested, and the intensity of the processed images was uniformly distributed throughout different brain areas. Furthermore, the processed image data set with its uniform brightness and high quality is now a fundamental atlas for image analysis, including cell segmentation, vascular tracing and visualization.