基于人工免疫及最大互信息的胸部多模图像配准*

为了实现胸部医学图像的自动配准，提出了一种基于人工免疫及最大互信息的配准方法。首先定义了最优配准的目标函数，接着运用人工免疫算法，结合最大互信息熵函数产生最优的仿射变换系数，从而实现医学图像自动配准。该人工免疫算法中，抗原是指最大互信息熵目标函数，而抗体是指最优的仿射变换系数。实验证明该方法配准效果较好。

Thorax multimodal image registration based on artificial immune algorithm and maximization of mutual information

In order to realize the automatic registration of pulmonary medical images, this paper proposed an image registration method based on artificial immune algorithm and maximization of mutual information. Firstly, presented the objective function for the optimal registration. Then,presented an artificial immune approach to automatically generate the optimal parameters of affine transform, thus realized the multimodal image registration. In this approach, regarded the objective function as antigens, and regarded the optimal parameters of affine transform as antibodies. Experiments demonstrate the good performance of the proposed method.