【摘要】 扩散加权图像的量化分析在临床诊断上有着广泛应用,而图像采集时病人呼吸、心脏运动导致的不同b值图像间的偏差对诊断结果有着重要影响,因此配准是精确量化估计的前提条件。由于由于不同b值的扩散加权图像中的信号衰减程度不同且同一b值图像内存在灰度不均匀性,因此使用传统的配准算法会导致在将不同b值图像配准到b0图像的过程中产生较大的偏移,尤其是在高b值的图像上。文中提出了拟合精度引导自由形变(Free-Form Deformation,FFD)模型的新方法,实现了多b值扩散加权图像的精确配准。所提方法应用体素不相干运动(Intra-Voxel Incoherent Motion,IVIM)模型对图像进行参数拟合从而得到拟合精度,并使用拟合精度构造的权重矩阵对图像中的不同位置自适应地加权自由形变的变形步长以得到最优的变形场。5组不同b值序列图像上的实验结果表明,所提方法提高了扩散加权图像的配准效果,且经过配准后获得了更加精确的IVIM模型参数。更多还原

【Abstract】 Quantifying analysis of diffusion-weighted images has been widely used in clinical diagnosis.But the misalignments among diffusion weighted images with different b values caused by patients’ respiratory and cardiac motion during scan lead to a reduced diagnostic value.Thus registration is a precondition for precise quantifying analysis.Conventional registration algorithms will cause unwanted distortions in the registration of different b-value images to b₀ image,especially for high-b-value images,due to the signal attenuation among different b-value images and the inhomogeneity in b₀ image.This paper proposed a new method that utilizes fitting accuracy to guide free-form deformation（FFD）model for precise registration of multi-b-value diffusion weighted images.Fitting accuracy was obtained by fitting DW images with intra-voxel incoherent motion（IVIM）model and then a weight matrix was constructed with fitting accuracy to adaptively weight step size in FFD for finding the optimal transformation field.The results of 5 series of multi-b-value diffusion weighted images indicate that the proposed method improves the registration property of diffusion weighted images and obtains more accurate IVIM parameters after registration.更多还原