Multimodal Medical Image Registration Based on an Information-Theory Measure with Histogram Estimation of Continuous Image Representation

Algorithm 2

Multimodel medical image registration.

Input: Reference image , floating image

Stage 1: Given the initialized transformation , Output is the transformed floating image .

Stage2: Random sampling strategy was used for and , and then output are the coordinates of sample points.

Stage 3: The joint histogram between and was estimated using these sample points. The output is the

joint probability distribution of and .

Stage 4: The joint probability distribution was adopted to calculate the JA similarity measure between and .

The output of this stage is the value of JA between and .

Stage 5: The simplex optimization method was employed. If the stop criteria of optimization method are satisfied,

output is the optimal transformation , otherwise go to stage 1.