Intensity based affine registration including feature similarity for spatial normalization.

摘要: This paper presents a new spatial normalization with affine transformation. The quantitative comparison of brain architecture across different subjects requires common coordinate system. For the analysis specific area, it is required to normalize and compare region interest global brain. Intensity based registration method matches well. But may not be locally normalized compared feature method. this uses similarities local as well intensity similarities. lateral ventricle central gray nuclei including corpus callosum, which used for features in Schizophrenia detection, appropriately normalized. In results section, our reduces difference area such callosum (7.7%,2.4%) (8.2%,13.5%) mutual information Talairach methods.