PET Image Reconstruction Based On Fuzzy Theory And Wavelet Transform

Positron Emission Tomography (PET) can provide tissue functional imaging with the aid of radioisotope, which currently makes it one of noninvasive medical imaging tools observing dynamically and quantitatively the physiological and biochemical changes in vivo. Two pivotal techniques in PET research area, image reconstruction and data acquisition, are gaining increasing importance and attention which are impulsed by the wide application of PET and fast development of computational techniques. Generally two reconstruction approaches are discussed in the paper: MAP reconstruction model based on fuzzy theory and anisotropic diffusion; and MAP reconstruction based on wavelet transform.In the first part of the paper, both fuzzy theory and anisotropic diffusion method are applied to PET reconstruction. The method of anisotropic diffusion is a superior method for image noise removing, which can remove the noise inside the image while keeping the edge of image. In this paper, the two theories are incorporated to image noise removing firstly, and then are applied to PET reconstruction. The feasibility and efficiency of the proposed method are verified experimentally in comparison with other commonly used methods.In the second part of the paper, PET reconstruction based on wavelet transform is thoroughly studied. The theory of wavelet transform is widely used in recent years, which has already shown its efficiency in the aspects of noise removing, image compression and so on. In this paper, wavelet transform is applied to PET reconstruction field. Firstly, the wavelet transform of the image is evaluated using the MAP method, and then the image can be evaluated by the back transform of wavelet. The simulation results show the proposed method can be superior to other MAP methods by providing high quality reconstructions with well preserved image details.