Fusion of CT and PET Medical Images Using Hybrid Algorithm DWT-DCT-PCA

The successful diagnosis of a disease depends on the accuracy of the image obtained from medical imaging modalities. Medical image fusion acts as a ´life saving tool´-thus it has emerged as a promising research field in recent years. The objective of medical imaging is to acquire a high resolution image with more information for the sake of diagnostic purposes. This paper proposes a hybrid fusion algorithm for multimodality medical images. There are two type of modalities one is ´Anatomy´, which gives structural details of body parts, such as X-ray, CT, MRI, and other ´Physiology and Metabolism´, it gives the information about functional details of cell activity in the organ, such as SPECT, PET. Structure without function is a corpse and function without structure is a ghost. Therefore, both of the Anatomy, Physiology and Metabolism images are investigated. So, this work makes fusion of CT and PET images. Specifically it aims at the gathering relevant, disparate and complementary data in one order to enhance the information apparent in the images, as well as to amplify the reliability of the interpretation. This leads to more accurate data and increased utility. In addition, it has been stated that combined data provides for robust operational performance such as increased confidence, reduced ambiguity, and improved reliability. This paper, introduced a pixel level based ´Hybrid Concept´ by integrating the conventional and advance fusion methods to overcome their demerits and enhance the image processing qualities like, PCA (Principal component analysis), DWT(Discrete Wavelet Transformation), DCT(Discrete Curvelet Transformation) to form a DWT-PCA, DCT-PCA, and proposing one DWT-DCT-PCA algorithm. These are analytically examined, observed and compared the results among them using the performance matrices MSE, PSNR and ENTROPY.