Post-Reconstruction Deconvolution of PET Images by Total Generalized Variation Regularization

TL;DR

This paper presents a novel PET image deconvolution method using total generalized variation regularization, improving image quality in clinical settings with noisy, low-resolution data, without requiring raw data access.

Contribution

Introduces TGV-based regularization for PET deconvolution, stabilized with positivity and photometry constraints, and an automatic regularization parameter adjustment for Poisson noise.

Findings

Enhanced PET image quality demonstrated on synthetic and real data.

Effective noise suppression and resolution improvement achieved.

Method applicable in clinical scenarios without raw data access.

Abstract

Improving the quality of positron emission tomography (PET) images, affected by low resolution and high level of noise, is a challenging task in nuclear medicine and radiotherapy. This work proposes a restoration method, achieved after tomographic reconstruction of the images and targeting clinical situations where raw data are often not accessible. Based on inverse problem methods, our contribution introduces the recently developed total generalized variation (TGV) norm to regularize PET image deconvolution. Moreover, we stabilize this procedure with additional image constraints such as positivity and photometry invariance. A criterion for updating and adjusting automatically the regularization parameter in case of Poisson noise is also presented. Experiments are conducted on both synthetic data and real patient images.