A robust state-space kinetics-guided framework for dynamic PET image reconstruction

Temporal information guided dynamic dual-tracer PET signal separation network

PURPOSE

The difficulty of dynamic dual-tracer positron emission tomography (PET) technology is to separate the complete single-tracer information from mixed dual-tracer. Traditional methods cannot separate single injection single-scan dynamic dual-tracer PET images. In this paper, we propose a deep learning framework based on gated recurrent unit (GRU) network and evaluate its performance with simulation experiments and realistic monkey data.

METHODS

The proposed single-scan dynamic dual-tracer PET image separation network consists of three parts, including encoder, separation and decoder module. Encoder part is to map the mixed time activity curves (TACs) from the low-dimensional space to the high-dimensional space to get mixed weight vector matrix. Separation part is to capture the temporal information of mixed weight vector matrix using bi-directional GRU (bi-GRU) layer to obtain the single-tracer masks, and the decoding part remaps the high-dimensional single-tracer weight vector matrix to the low-dimensional space to obtain two separated single tracers.

RESULTS

In the simulation experiments under different tracers, phantoms, noise levels, arterial input function (AIF) and k-parameter with Gaussian random, compared to the stacked auto encoder (SAE) network and traditional background subtraction method, GRU-based network has better performance with low bias and mean squared error (MSE). In the realistic study, the image results of GRU network have higher mean structural similarity (MSSIM), and peak signal to noise ratio (PSNR).

CONCLUSIONS

This study demonstrates the feasibility of temporal information guided neural network in single-injection single-scan dynamic dual-tracer PET images separation. The GRU-based network uses TAC temporal information without AIFs to make the separation results more robust and accurate, which significantly outperforms state-of-the-art method qualitatively and quantitatively. This article is protected by copyright. All rights reserved.

Dynamic PET reconstruction using temporal patch-based low rank penalty for ROI-based brain kinetic analysis

Dynamic positron emission tomography (PET) is widely used to measure changes in the bio-distribution of radiopharmaceuticals within particular organs of interest over time. However, to retain sufficient temporal resolution, the number of photon counts in each time frame must be limited. Therefore, conventional reconstruction algorithms such as the ordered subset expectation maximization (OSEM) produce noisy reconstruction images, thus degrading the quality of the extracted time activity curves (TACs). To address this issue, many advanced reconstruction algorithms have been developed using various spatio-temporal regularizations. In this paper, we extend earlier results and develop a novel temporal regularization, which exploits the self-similarity of patches that are collected in dynamic images. The main contribution of this paper is to demonstrate that the correlation of patches can be exploited using a low-rank constraint that is insensitive to global intensity variations. The resulting optimization framework is, however, non-Lipschitz and nonconvex due to the Poisson log-likelihood and low-rank penalty terms. Direct application of the conventional Poisson image deconvolution by an augmented Lagrangian (PIDAL) algorithm is, however, problematic due to its large memory requirements, which prevents its parallelization. Thus, we propose a novel optimization framework using the concave-convex procedure (CCCP)

Graphics processing unit (GPU)-accelerated particle filter framework for positron emission tomography image reconstruction

As a consequence of the random nature of photon emissions and detections, the data collected by a positron emission tomography (PET) imaging system can be shown to be Poisson distributed. Meanwhile, there have been considerable efforts within the tracer kinetic modeling communities aimed at establishing the relationship between the PET data and physiological parameters that affect the uptake and metabolism of the tracer. Both statistical and physiological models are important to PET reconstruction. The majority of previous efforts are based on simplified, nonphysical mathematical expression, such as Poisson modeling of the measured data, which is, on the whole, completed without consideration of the underlying physiology. In this paper, we proposed a graphics processing unit (GPU)-accelerated reconstruction strategy that can take both statistical model and physiological model into consideration with the aid of state-space evolution equations. The proposed strategy formulates the organ activity distribution through tracer kinetics models and the photon-counting measurements through observation equations, thus making it possible to unify these two constraints into a general framework. In order to accelerate reconstruction, GPU-based parallel computing is introduced. Experiments of Zubal-thorax-phantom data, Monte Carlo simulated phantom data, and real phantom data show the power of the method. Furthermore, thanks to the computing power of the GPU, the reconstruction time is practical for clinical application.