3D Multi-Modality Medical Image Registration based on Quantum-behaved Particle Swarm Optimization Algorithm

Image registration based on mutual information is of high accuracy and robustness. Unfortunately, the mutual information function is generally not a smooth function but one containing many local maxima, which has a large influence on optimization. This paper proposes a registration method based on Quantum-behaved Particle Swarm Optimization Algorithm. Not only QPSO has less parameters to con-troll, but also does its sampling space at each iteration covers the whole solution space. Thus QPSO can find the best solution quickly and guarantee to be global convergent. Experiments shows that this registration method could efficiently restrain local maxima of mutual information function and it can improve accuracy. Compare with the gold standard, the subvoxel accuracy can be achieved.