【摘要】 在 3 D多模医学图像的配准方法中 ,最大互信息法精度高、鲁棒性强、使用范围广。本文将归一化互信息作为相似性测度 ,采用不同的采样范围和采样子集 ,使用 Powell多参数优化法和 Brent一维搜索算法对 3 D CT、MR和 PET脑图像进行了刚体配准。为了加快配准速度 ,使用了多分辨的金字塔方法。对 PET图像采用基于坐标的阈值选取方法对图像进行分割预处理 ,消除了大部分放射状背景伪影。美国万德贝尔大学对结果进行的评估证明配准精度可达亚体元级更多还原

【Abstract】 Maximization of mutual information is a powerful criterion for 3D medical image registration, allowing robust and fully accurate automated rigid registration of multi-modal images in a various applications. In this paper, a method based on normalized mutual information for 3D image registration was presented on the images of CT, MR and PET. Powell’s direction set method and Brent’s one-dimensional optimization algorithm were used as optimization strategy. A multi-resolution approach is applied to speedup the matching process. For PET images, pre-procession of segmentation was performed to reduce the background artefacts. According to the evaluation by the Vanderbilt University, Sub-voxel accuracy in multi-modality registration had been achieved with this algorithm.更多还原