PET image was often influenced by noise. In this paper, we proposed a nonlinear variational model for improving reconstruction of PET images. The use of variational model was due to its effectiveness for reducing noise in 2D images while preserving edges. Our results indicated that the proposed method application to computer-simulated and real PET phantom outperformed the conventional method in terms of both visual quality and quantitative accuracy.
Citation:
Jianhua Yan, Jun Yu, "A Nonlinear Variational Model for PET Reconstruction," icpr, vol. 4, pp.699-702, 18th International Conference on Pattern Recognition (ICPR'06) Volume 4, 2006