Dynamic positron emission tomography image restoration via a kinetics-induced bilateral filter

Dose reduction in dynamic synaptic vesicle glycoprotein 2A PET imaging using artificial neural networks

Objective. Reducing dose in positron emission tomography (PET) imaging increases noise in reconstructed dynamic frames, which inevitably results in higher noise and possible bias in subsequently estimated images of kinetic parameters than those estimated in the standard dose case. We report the development of a spatiotemporal denoising technique for reduced-count dynamic frames through integrating a cascade artificial neural network (ANN) with the highly constrained back-projection (HYPR) scheme to improve low-dose parametric imaging.Approach. We implemented and assessed the proposed method using imaging data acquired with11C-UCB-J, a PET radioligand bound to synaptic vesicle glycoprotein 2A (SV2A) in the human brain. The patch-based ANN was trained with a reduced-count frame and its full-count correspondence of a subject and was used in cascade to process dynamic frames of other subjects to further take advantage of its denoising capability. The HYPR strategy was then applied to the spatial ANN processed image frames to make use of the temporal information from the entire dynamic scan.Main results. In all the testing subjects including healthy volunteers and Parkinson's disease patients, the proposed method reduced more noise while introducing minimal bias in dynamic frames and the resulting parametric images, as compared with conventional denoising methods.Significance. Achieving 80% noise reduction with a bias of -2% in dynamic frames, which translates into 75% and 70% of noise reduction in the tracer uptake (bias, -2%) and distribution volume (bias, -5%) images, the proposed ANN+HYPR technique demonstrates the denoising capability equivalent to a 11-fold dose increase for dynamic SV2A PET imaging with11C-UCB-J.

The Consistence of Dynamic Contrast-Enhanced MRI and Filter-Exchange Imaging in Measuring Water Exchange Across the Blood-Brain Barrier in High-Grade Glioma

BACKGROUND

Water exchange across blood-brain barrier (BBB) (WEXBBB ) is an emerging biomarker of BBB dysfunction with potential applications in many brain diseases. Several MRI methods have been proposed to measure WEXBBB , but evidence remains scarce whether different methods can produce comparable WEXBBB .

PURPOSE

To explore whether dynamic contrast-enhanced (DCE)-MRI and vascular water exchange imaging (VEXI) could produce comparable WEXBBB in high-grade glioma (HGG) patients.

STUDY TYPE

Prospective cross-sectional.

SUBJECTS

13 HGG patients (58.4 ± 9.4 years, 9 females, 4 WHO III and 9 WHO IV).

FIELD STRENGTH/SEQUENCE

A 3 T, spoiled gradient-recalled-echo DCE-MRI and VEXI containing two pulsed-gradient spin-echo blocks separated by a mixing block.

ASSESSMENTS

The enhanced tumor and contralateral normal-appearing white matter (cNAWM) volume-of-interests (VOIs) were drew by two neuroradiologists. And whole-brain NAWM and normal-appearing gray matter (NAGM) without tumor-affected regions were segmented by automated segmentation algorithm in FSL.

STATISTICAL TESTS

Student's t-test was used to evaluate parameters difference between cNAWM and tumor, NAGM and NAWM, respectively. The correlation between vascular water efflux rate constant (kbo ) from DCE-MRI and apparent exchange rate across BBB (AXRBBB ) from VEXI was evaluated by Pearson correlation. P < 0.05 was considered statistically significant.

RESULTS

Compared with cNAWM, both kbo and AXRBBB were significantly reduced in tumor (kbo  = 3.50 ± 1.18 sec-1 vs. 1.03 ± 0.75 sec-1 ; AXRBBB  = 3.54 ± 1.11 sec-1 vs. 1.94 ± 1.04 sec-1 ). Both kbo and AXRBBB showed significantly higher values in NAWM than NAGM (kbo  = 3.50 ± 0.59 sec-1 vs. 2.10 ± 0.56 sec-1 ; AXRBBB  = 3.35 ± 0.77 sec-1 vs. 2.07 ± 0.52 sec-1 ). The VOI-averaged kbo and AXRBBB were also linearly correlated in tumor, NAWM, and NAGM (r = 0.59).

DATA CONCLUSION

DCE-MRI and VEXI showed comparable and correlated WEXBBB in HGG patients, suggesting that the consistence and reliability of these two MRI methods in measuring WEXBBB .

EVIDENCE LEVEL

2.

TECHNICAL EFFICACY

Stage 1.

Restoration of Tc-99m methyl diphosphonate bone scan image using Richardson-Lucy algorithm

INTRODUCTION

In this study, the optimal input parameters point spread function (PSF) and the number of iterations of the Richardson-Lucy algorithm were experimentally determined to restore Tc-99 m methyl diphosphonate (MDP) whole-body bone scan images.

MATERIALS AND METHODS

The experiment was performed on 60 anonymized Tc-99 m MDP whole-body bone scan images. Ten images were used for estimating the optimum value of PSF and the number of iterations to restore scintigraphic images. The remaining 50 images were used for validation of estimated parameters. The image quality of observed and restored images was assessed objectively using blind/referenceless image spatial quality evaluator (BRISQUE), mean brightness (MB), discrete entropy (DE), and edge-based contrast measure (EBCM) image quality metrics. Image quality was subjectively assessed by two nuclear medicine physicians (NMPs) by comparing the restored image quality with observed image quality and assigning a score to each image on the scale of 0-5.

RESULTS

Based on BRISQUE, MB, DE, and EBCM scores, the restored images were significantly sharper, less bright, had more detailed information, and had less contrast around edges compared to the input images. The restored images had improved resolution based on visual assessment as well; NMPs assigned an average image quality score of 4.00 to restored images. Maximum resolution enhancement was noticed at PSF (size: 11 pixels, sigma: 1.75 pixels) and the number of iterations = 10. With the increase in the number of iterations, noise also gets amplified along with resolution enhancement and affects the detectability of small lesions; in the case of relatively low noisy input images, the number of iterations = 5 gave better results.

CONCLUSION

Tc-99 m MDP bone scan images were restored to improve image quality using the Richardson-Lucy algorithm. The optimum value of the PSF parameter was found to be of size = 11 pixels and sigma = 1.75 pixels.

Noninvasive quantification of nonhuman primate dynamic 18F-FDG PET imaging

¹⁸F-FDG uptake rate constant K_i is the main physiology parameter measured in dynamic PET studies. A model-independent graphical analysis using Patlak plot with plasma input function (PIF) is a standard approach used to estimate K_i . The PIF is the ¹⁸F-FDG time activity curve (TAC) in plasma that is obtained by serial arterial blood sampling. The purpose of the study is to evaluate a Patlak plot-based optimization approach with reduced blood samples for noninvasive quantification of dynamic ¹⁸F-FDG PET imaging. Eight 60 min rhesus monkey brain dynamic ¹⁸F-FDG PET scans with arterial blood samples were collected. The measured PIF (mPIF) was determined by arterial blood samples. TACs of seven cerebral regions of interest were generated from each study. With a given number of blood samples, the population-based PIF (pPIF) was determined by either interpolation or extrapolation method using scale calibrated population mean of normalized PIF. The optimal sampling scheme with given blood sample size was determined by maximizing the correlations between the K_i estimated from pPIF and those obtained by mPIF. A leave-two-out cross-validation method was used for evaluation. The linear correlations between the K_i estimates from pPIF with optimal sampling schemes and those from mPIF were: K_i (pPIF 1 sample at 40 min) = 1.015 K_i (mPIF) - 0.000, R ² = 0.974; K_i (pPIF 2 samples at 35 and 50 min) = 1.052 K_i (mPIF) - 0.001, R ² = 0.976; K_i (pPIF 3 samples at 12, 40, and 50 min) = 1.030 K_i (mPIF) - 0.000, R ² = 0.985; and K_i (pPIF 4 samples at 10, 20, 40, and 50 min) = 1.016 K_i (mPIF)- 0.000, R ² = 0.993. As the sample size became greater or equal to 4, the K_i estimates from pPIF with the optimal protocol were almost identical to those from mPIF. The Patlak plot-based optimization approach is a reliable method to estimate PIF for noninvasive quantification of non-human primate dynamic ¹⁸F-FDG PET imaging and is potentially extendable to further translational human studies.

Denoising PET images for proton therapy using a residual U-net

The use of proton therapy has the advantage of high dose concentration as it is possible to concentrate the dose on the tumor while suppressing damage to the surrounding normal organs. However, the range uncertainty significantly affects the actual dose distribution in the vicinity of the proton range, limiting the benefit of proton therapy for reducing the dose to normal organs. By measuring the annihilation gamma rays from the produced positron emitters, it is possible to obtain a proton induced positron emission tomography (pPET) image according to the irradiation region of the proton beam. Smoothing with a Gaussian filter is generally used to denoise PET images; however, this approach lowers the spatial resolution. Furthermore, other conventional smoothing processing methods may deteriorate the steep region of the pPET images. In this study, we proposed a denoising method based on a Residual U-Net for pPET images. We conducted the Monte Carlo simulation and irradiation experiment on a human phantom to obtain pPET data. The accuracy of the range estimation and the image similarity were evaluated for pPET images using the Residual U-Net, a Gaussian filter, a median filter, the block-matching and 3D-filtering (BM3D), and a total variation (TV) filter. Usage of the Residual U-Net yielded effective results corresponding to the range estimation; however, the results of peak-signal-to-noise ratio were identical to those for the Gaussian filter, median filter, BM3D, and TV filter. The proposed method can contribute to improving the accuracy of treatment verification and shortening the PET measurement time.

LCPR-Net: low-count PET image reconstruction using the domain transform and cycle-consistent generative adversarial networks

BACKGROUND

Reducing the radiation tracer dose and scanning time during positron emission tomography (PET) imaging can reduce the cost of the tracer, reduce motion artifacts, and increase the efficiency of the scanner. However, the reconstructed images to be noisy. It is very important to reconstruct high-quality images with low-count (LC) data. Therefore, we propose a deep learning method called LCPR-Net, which is used for directly reconstructing full-count (FC) PET images from corresponding LC sinogram data.

METHODS

Based on the framework of a generative adversarial network (GAN), we enforce a cyclic consistency constraint on the least-squares loss to establish a nonlinear end-to-end mapping process from LC sinograms to FC images. In this process, we merge a convolutional neural network (CNN) and a residual network for feature extraction and image reconstruction. In addition, the domain transform (DT) operation sends a priori information to the cycle-consistent GAN (CycleGAN) network, avoiding the need for a large amount of computational resources to learn this transformation.

RESULTS

The main advantages of this method are as follows. First, the network can use LC sinogram data as input to directly reconstruct an FC PET image. The reconstruction speed is faster than that provided by model-based iterative reconstruction. Second, reconstruction based on the CycleGAN framework improves the quality of the reconstructed image.

CONCLUSIONS

Compared with other state-of-the-art methods, the quantitative and qualitative evaluation results show that the proposed method is accurate and effective for FC PET image reconstruction.

4D deep image prior: dynamic PET image denoising using an unsupervised four-dimensional branch convolutional neural network

Although convolutional neural networks (CNNs) demonstrate the superior performance in denoising positron emission tomography (PET) images, a supervised training of the CNN requires a pair of large, high-quality PET image datasets. As an unsupervised learning method, a deep image prior (DIP) has recently been proposed; it can perform denoising with only the target image. In this study, we propose an innovative procedure for the DIP approach with a four-dimensional (4D) branch CNN architecture in end-to-end training to denoise dynamic PET images. Our proposed 4D CNN architecture can be applied to end-to-end dynamic PET image denoising by introducing a feature extractor and a reconstruction branch for each time frame of the dynamic PET image. In the proposed DIP method, it is not necessary to prepare high-quality and large patient-related PET images. Instead, a subject's own static PET image is used as additional information, dynamic PET images are treated as training labels, and denoised dynamic PET images are obtained from the CNN outputs. Both simulation with [¹⁸F]fluoro-2-deoxy-D-glucose (FDG) and preclinical data with [¹⁸F]FDG and [¹¹C]raclopride were used to evaluate the proposed framework. The results showed that our 4D DIP framework quantitatively and qualitatively outperformed 3D DIP and other unsupervised denoising methods. The proposed 4D DIP framework thus provides a promising procedure for dynamic PET image denoising.

A kernel-based image denoising method for improving parametric image generation

One of the main challenges in the pixel-wise modeling analysis is the presence of high noise levels. Wang and Qi proposed a kernel-based method for dynamic positron emission tomgraphy reconstruction. Inspired by this method, we propose a kernel-based image denoising method based on the minimization of a kernel-based lp-norm regularized problem. To solve the kernel-based image denoising problem, we used the general-threshold filtering algorithm in combination with total difference. In the present study, we investigated whether diffusion-weighted magnetic resonance imaging (DW-MRI) data denoised using the proposed method can provide improved intravoxel incoherent motion (IVIM) parametric images. We also compared the proposed method with the method using the local principal component analysis (LPCA). The simulated DW-MR magnitude images are assumed to have Rician distributed noise. Computer simulations show that the proposed image denoising method can achieve a better bias-variance trade-off than the LPCA method. Moreover, the proposed method can reduce variance while simultaneously preserving edges in the parametric images. We tested our image denoising method on in vivo DW-MRI data, and the result showed that the denoised DWI-MRI data obtained using the proposed method can substantially improve the quality of IVIM parametric images.

Low-dose dynamic myocardial perfusion CT imaging using a motion adaptive sparsity prior

PURPOSE

Dynamic myocardial perfusion computed tomography (DM-PCT) imaging offers benefits over quantitative assessment of myocardial blood flow (MBF) for diagnosis and risk stratification of coronary artery disease. However, one major drawback of DM-PCT imaging is that a high radiation level is imparted by repeated scanning. To address this issue, in this work, we developed a statistical iterative reconstruction algorithm based on the penalized weighted least-squares (PWLS) scheme by incorporating a motion adaptive sparsity prior (MASP) model to achieve high-quality DM-PCT imaging with low tube current dynamic data acquisition. For simplicity, we refer to the proposed algorithm as "PWLS-MASP''.

METHODS

The MASP models both the spatial and temporal structured sparsity of DM-PCT sequence images with the assumption that the differences between adjacent frames after motion correction are sparse in the gradient image domain. To validate and evaluate the effectiveness of the present PWLS-MASP algorithm thoroughly, a modified XCAT phantom and preclinical porcine DM-PCT dataset were used in the study.

RESULTS

The present PWLS-MASP algorithm can obtain high-quality DM-PCT images in both phantom and porcine cases, and outperforms the existing filtered back-projection algorithm and PWLS-based algorithms with total variation regularization (PWLS-TV) and robust principal component analysis regularization (PWLS-RPCA) in terms of noise reduction, streak artifacts mitigation, and time density curve estimation. Moreover, the PWLS-MASP algorithm can yield more accurate diagnostic hemodynamic parametric maps than the PWLS-TV and PWLS-RPCA algorithms.

CONCLUSIONS

The study indicates that there is a substantial advantage in using the present PWLS-MASP algorithm for low-dose DM-PCT, and potentially in other dynamic tomography areas.

Dynamic positron emission tomography restoration with low-rank representation incorporating edge preservation

BACKGROUND

Dynamic positron emission tomography (PET) is a powerful tool that provides useful quantitative information on physiological and biochemical processes. However, the low signal-to-noise ratio (SNR) in short dynamic frames is a challenge.

OBJECTIVE

To get high SNR in the dynamic PET and to achieve high-quality PET parametric image are the objective of this study.

METHODS

Low-rank (LR) modeling and edge-preserving prior are incorporated in this study with a unified mathematical framework to improve the SNR of a dynamic PET image series. The proposed algorithm is designed to reduce noise in homogeneous areas while preserving the edges of regions of interest.

RESULTS

The performance of the proposed method (LRH) is compared both visually and quantitatively by using the classic Gaussian filter and an LR expression filter on a digital brain phantom and in vivo rat study. Experimental results demonstrate that the proposed filter can achieve superior visual and quantitative performance without sacrificing spatial resolution.

CONCLUSIONS

The proposed LRH is considerably effective and exhibits great potential in processing dynamic PET data with high noise levels.

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)