Miao Peng, Li Nan, Rege Abhishek, Tong Shanbao, Thakor Nitish
Med-X Research Institute, Shanghai Jiao Tong University, 1954 Huashan Road, Shanghai 200030, China.
Annu Int Conf IEEE Eng Med Biol Soc. 2009;2009:3236-93. doi: 10.1109/IEMBS.2009.5333603.
Laser speckle imaging (LSI) and optical intrinsic signal (OIS) imaging are shown to produce high resolution information of cerebral blood flow(CBF) and de-oxygen hemoglobin(dHb) respectively. However, the cortex curvature and non-uniform illumination always result in the inhomo-geneous impact and thus distort the CBF and dHb results. In this paper, we propose to extract both CBF and dHb images from raw speckle images. A parametric model for such inhomogeneity is estimated by adaptive window median-mean filtering and curve fitting technique to reconstruct the CBF and dHb image.
激光散斑成像（LSI）和光学内在信号（OIS）成像分别被证明可产生脑血流量（CBF）和脱氧血红蛋白（dHb）的高分辨率信息。然而，皮质曲率和光照不均匀总是导致不均匀的影响，从而扭曲CBF和dHb的结果。在本文中，我们建议从原始散斑图像中提取CBF和dHb图像。通过自适应窗口中值均值滤波和曲线拟合技术估计这种不均匀性的参数模型，以重建CBF和dHb图像。
