New Anatomical-Prior-Based Image Reconstruction Method for PET/SPECT

摘要

We propose a new anatomical-prior-based reconstruction method for PET/SPECT. The method first creates a template image simulating a PET/SPECT image from an anatomical image, and incorporates a distance between the template image and the reconstructed image into the cost function for image reconstruction. The most significant feature of the proposed method is that the distance between the template image and the reconstructed image is evaluated by using L1/L0 norm distance functions. We demonstrate that the use of L1/L0 norm distances possesses an incredibly strong power in improving the reconstructed image.