提出了一种新的基于等效子午面和互信息量的三维医学图像快速配准算法——EMP—MI算法.传统的互信息量的方法需要考虑整个三维数据的信息,计算复杂度大,无法满足临床需要.而本算法将三维数据的配准转化为二维数据的配准,在保证精度的前提下,减少了配准所需时间.文中创新点在于利用主成分分析计算出图像的等效子午面并将图像转化到标准坐标系下,从而将质心和等效子午面粗配准,精细配准时只需要对浮动图像进行微小的调整计算等效子午面的互信息量,这就大大提高了配准速度,减少了陷入局部极值的可能.实验结果表明这种先整体后局部的方法能准确、快速地处理图像刚性配准问题,特别适用于三维医学图像的配准.
This paper presents a new robust and fast 3-D image registration method based on the equivalent meridian plane (EMP) and mutual information (MI). Comparing with traditional MI based registration methods that estimate the MI using the whole volume intensity information, our approach uses the 2D plane--Equivalent meridian plane's information. A novel aspect of our approach is using principal component analysis to find the equivalent meridian plane, and then compute its MI. We evaluate the effectiveness of the EMP-MI approach by applying it to the simulated and real brain image data. The experimental results indicate that the algorithm is effective in reducing computation time as well as in helping to avoid local minima.