Faster PET Reconstruction with Non-Smooth Priors by Randomization and Preconditioning

Abstract

Uncompressed clinical data from modern positron emission tomography (PET) scanners are very large, exceeding 350 million data points (projection bins). The last decades have seen tremendous advancements in mathematical imaging tools many of which lead to non-smooth (i.e. non-differentiable) optimization problems which are much harder to solve than smooth optimization problems. Most of these tools have not been translated to clinical PET data, as the state-of-the-art algorithms for non-smooth problems do not scale well to large data. In this work, inspired by big data machine learning applications, we use advanced randomized optimization algorithms to solve the PET reconstruction problem for a very large class of non-smooth priors which includes for example total variation, total generalized variation, directional total variation and various different physical constraints. The proposed algorithm randomly uses subsets of the data and only updates the variables associated with these. While this idea often leads to divergent algorithms, we show that the proposed algorithm does indeed converge for any proper subset selection. Numerically, we show on real PET data (FDG and florbetapir) from a Siemens Biograph mMR that about ten projections and backprojections are sufficient to solve the MAP optimisation problem related to many popular non-smooth priors; thus showing that the proposed algorithm is fast enough to bring these models into routine clinical practice.

References in corpus (9)

• Convex optimization problem prototyping for image reconstruction in computed tomography with the Chambolle-Pock algorithm
• This is SPIRAL-TAP: Sparse Poisson Intensity Reconstruction ALgorithms - Theory and Practice
• An algorithm for constrained one-step inversion of spectral CT data
• Joint Reconstruction of Multi-channel, Spectral CT Data via Constrained Total Nuclear Variation Minimization
• Multi-Contrast MRI Reconstruction with Structure-Guided Total Variation
• Blind Image Fusion for Hyperspectral Imaging with the Directional Total Variation
• A function space framework for structural total variation regularization with applications in inverse problems
• A Framework for Directional and Higher-Order Reconstruction in Photoacoustic Tomography
• Joint Reconstruction via Coupled Bregman Iterations with Applications to PET-MR Imaging

Cited by in corpus (8)

• Parameter-Transferred Wasserstein Generative Adversarial Network (PT-WGAN) for Low-Dose PET Image Denoising
• Core Imaging Library -- Part II: Multichannel reconstruction for dynamic and spectral tomography
• The Practicality of Stochastic Optimization in Imaging Inverse Problems
• Robust Image Reconstruction with Misaligned Structural Information
• Multi-modality imaging with structure-promoting regularisers
• List-Mode PET Image Reconstruction Using Dykstra-Like Splitting
• Stochastic EM methods with Variance Reduction for Penalised PET Reconstructions
• Fast and memory-efficient reconstruction of sparse Poisson data in listmode with non-smooth priors with application to time-of-flight PET