PRINCIPAL COMPONENT ANALYSIS OF DYNAMIC POSITRON EMISSION TOMOGRAPHY IMAGES

Multivariate image analysis can be used to analyse multivariate medica l images. The purpose could be to visualize or classify structures in the image. One common multivariate image analysis technique which can be used for visualization purposes is principal component analysis (PC A). The present work concerns visualization of organs and structures w ith different kinetics in a dynamic sequence utilizing PCA. When apply ing PCA on positron emission tomography (PET) images, the result is in itially not satisfactory. It is illustrated that one major explanation for the behaviour of PCA when applied to PET images is that it is a d ata-driven technique which cannot separate signals from high noise lev els, With a better understanding of the PCA, gained with a strategy of examining the image data set, the transformations, and the results us ing visualization tools, a surprisingly easily understood be derived. The proposed enhance clinically interesting information in a dynamic P ET imaging sequence in the first few principal component images and th us should be able to aid in the identification of structures for furth er analysis.