基于互信息和混合优化算法的多模医学图像配准

Multimodality Medical Image Registration Based on Mutual Information and Hybrid Optimization Algorithm

针对互信息函数的多极值问题,提出了一种基于混合优化算法的多模医学图像配准方法.对于多模医学图像,以互信息作为相似性测度,使用混合优化算法搜索出最佳配准变换参数,将待配准图像进行变换,从而达到配准的目的.实验表明,该算法能避免陷入局部最优值,配准结果精度达到亚像素级.

A multimodality medical image registration method based on hybrid optimization algorithm was proposed to solve the problem of the mutual information function containing many local maxima. When dealing with multimodality medical images, we investigated the best matching parameters by applying mutual information as the similarity measure and hybrid optimization algorithm as the search strategy. The registration results showed that the subvoxel accuracy could be achieved and this method can avoid getting into the local optimum.