Nonlinear Diffusion Filter for Low Count Positron Emission Tomography Utilizing Orientation Information of Neighbouring Gradient Vectors

Abstract

Positron Emission Tomography (PET) data are inherently noisy due to low photon counts. To achieve an acceptable signal-to-noise ratio (SNR), Gaussian Filter (GF) with varying levels of full width half maximum (FWHM) accounting for the intrinsic resolutions of different scanners are employed at the expense of reduction in spatial resolution and hence accurate delineation and quantification of lesions. Edge preserving Bilateral Filter (BF) provides a potential alternative to GF along with Nonlinear Diffusion Filter (NDF) with limited applicability for high count and low noise cases only. A novel parameter free nonlinear diffusion filter using gradient vector orientation information for clinical low count and high noise case is proposed in this paper and results are compared with GF, BF and NDF for different count levels for NEMA torso phantom.