Memory-Efficient Training for Fully Unrolled Deep Learned PET Image Reconstruction with Iteration-Dependent Targets

Comparison of 18F-FDG PET image quality and quantitative parameters between DPR and OSEM reconstruction algorithm in patients with lung cancer

OBJECTIVES

The present study aimed to investigate the influence of the deep progressive learning reconstruction (DPR) algorithm on the 18F-FDG PET image quality and quantitative parameters.

METHODS

In this retrospective study, data were collected from 55 healthy individuals and 184 patients with primary malignant pulmonary tumors who underwent 18F-FDG PET/CT examinations. PET data were reconstructed using the ordered subset expectation maximization (OSEM) and DPR algorithms. The influence of DPR algorithm on quantitative parameters was explored, including the SUVmax, SUVmean, standard deviation of SUV (SUVSD), metabolic tumor volume (MTV), total lesion glycolysis (TLG), and tumor-to-background uptake ratio (TBR). Finally, the differences in image quality parameters, including signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR), between the two reconstruction algorithms were evaluated.

RESULTS

DPR algorithm significantly reduced the SUVmax and SUVSD of background tissues (all, P < 0.001) compared to OSEM algorithm, while no statistical difference was observed in SUVmean between the two algorithms (all, P > 0.05). DPR algorithm notably increased the SUVmax, SUVmean, and TBR of lesions (all, P < 0.001) and reduced MTV (P = 0.005), with minimal differences in TLG noted between the reconstruction algorithms (P < 0.001). The percentage differences in SUVmax (P = 0.001), SUVmean (P = 0.005), and TBR (P = 0.001) between the two algorithms were significantly higher in solid nodules than in pure ground glass nodules (pGGNs). The ΔCNR between solid nodules (P = 0.031) and mixed ground glass nodules (P = 0.020) was greater than that between pGGNs. SNR and CNR obtained using the DPR algorithm were markedly improved compared to those determined using the OSEM algorithm (all, P < 0.001).

CONCLUSION

Under identical acquisition conditions, the DPR algorithm enhanced the accuracy of quantitative parameters in pulmonary lesions and potentially improved lesion detectability. The DPR algorithm increased image SNR and CNR compared to those obtained using the OSEM algorithm, significantly optimizing overall image quality. This advancement facilitated precise clinical diagnosis, underpinning its potential to significantly contribute to the field of medical imaging.

Clinical and Deep-Learned Evaluation of MR-Guided Self-Supervised PET Reconstruction

Reduced dose Positron Emission Tomography (PET) lowers the radiation dose to patients and reduces costs. Lower count data, however, degrades reconstructed image quality. Advanced reconstruction methods help mitigate image quality losses, but it is important to assess the resulting images from a clinical perspective. Two experienced clinicians assessed four PET reconstruction algorithms for [18F]FDG brain data, compared to a clinical standard reference (Maximum-Likelihood Expectation-Maximization (MLEM)), based on seven clinical image quality metrics: global quality rating, pattern recognition, diagnostic confidence (all on a scale of 0-4), sharpness, caudate-putamen separation, noise, and contrast (on a scale between 0-2). The reconstruction methods assessed were a guided and unguided version of self-supervised maximum a posteriori EM (MAPEM) (where the guidance case used the patient's MR image to control the smoothness penalty). For 3 of the 11 patient datasets reconstructed, post-smoothed versions of the MAPEM reconstruction were also considered, where the smoothing was with the point-spread-function used in the resolution modelling. Statistically significant improvements were observed in sharpness, caudate-putamen separation, and contrast for self-supervised MR-guided MAPEM compared to MLEM. For example, MLEM scored between 1-1.1 out of 2 for sharpness, caudate-putamen separation and contrast, whereas self-supervised MR-guided MAPEM scored between 1.5-1.75. In addition to the clinical evaluation, pre-trained Convolutional Neural Networks (CNNs) were used to assess the image quality of a further 62 images. The CNNs demonstrated similar trends to the clinician, showing their potential as automated standalone observers. Both the clinical and CNN assessments suggest when using only 5% of the standard injected dose, self-supervised MR-guided MAPEM reconstruction matches the 100% MLEM case for overall performance. This makes the images far more clinically useful than standard MLEM.

MPGAN: Multi Pareto Generative Adversarial Network for the denoising and quantitative analysis of low-dose PET images of human brain

Positron emission tomography (PET) imaging is widely used in medical imaging for analyzing neurological disorders and related brain diseases. Usually, full-dose imaging for PET ensures image quality but raises concerns about potential health risks of radiation exposure. The contradiction between reducing radiation exposure and maintaining diagnostic performance can be effectively addressed by reconstructing low-dose PET (L-PET) images to the same high-quality as full-dose (F-PET). This paper introduces the Multi Pareto Generative Adversarial Network (MPGAN) to achieve 3D end-to-end denoising for the L-PET images of human brain. MPGAN consists of two key modules: the diffused multi-round cascade generator (GDmc) and the dynamic Pareto-efficient discriminator (DPed), both of which play a zero-sum game for n(n∈1,2,3) rounds to ensure the quality of synthesized F-PET images. The Pareto-efficient dynamic discrimination process is introduced in DPed to adaptively adjust the weights of sub-discriminators for improved discrimination output. We validated the performance of MPGAN using three datasets, including two independent datasets and one mixed dataset, and compared it with 12 recent competing models. Experimental results indicate that the proposed MPGAN provides an effective solution for 3D end-to-end denoising of L-PET images of the human brain, which meets clinical standards and achieves state-of-the-art performance on commonly used metrics.

Single-Subject Deep-Learning Image Reconstruction With a Neural Optimization Transfer Algorithm for PET-Enabled Dual-Energy CT Imaging

Combining dual-energy computed tomography (DECT) with positron emission tomography (PET) offers many potential clinical applications but typically requires expensive hardware upgrades or increases radiation doses on PET/CT scanners due to an extra X-ray CT scan. The recent PET-enabled DECT method allows DECT imaging on PET/CT without requiring a second X-ray CT scan. It combines the already existing X-ray CT image with a 511 keV γ -ray CT (gCT) image reconstructed from time-of-flight PET emission data. A kernelized framework has been developed for reconstructing gCT image but this method has not fully exploited the potential of prior knowledge. Use of deep neural networks may explore the power of deep learning in this application. However, common approaches require a large database for training, which is impractical for a new imaging method like PET-enabled DECT. Here, we propose a single-subject method by using neural-network representation as a deep coefficient prior to improving gCT image reconstruction without population-based pre-training. The resulting optimization problem becomes the tomographic estimation of nonlinear neural-network parameters from gCT projection data. This complicated problem can be efficiently solved by utilizing the optimization transfer strategy with quadratic surrogates. Each iteration of the proposed neural optimization transfer algorithm includes: PET activity image update; gCT image update; and least-square neural-network learning in the gCT image domain. This algorithm is guaranteed to monotonically increase the data likelihood. Results from computer simulation, real phantom data and real patient data have demonstrated that the proposed method can significantly improve gCT image quality and consequent multi-material decomposition as compared to other methods.

AI for PET image reconstruction

Image reconstruction for positron emission tomography (PET) has been developed over many decades, with advances coming from improved modelling of the data statistics and improved modelling of the imaging physics. However, high noise and limited spatial resolution have remained issues in PET imaging, and state-of-the-art PET reconstruction has started to exploit other medical imaging modalities (such as MRI) to assist in noise reduction and enhancement of PET's spatial resolution. Nonetheless, there is an ongoing drive towards not only improving image quality, but also reducing the injected radiation dose and reducing scanning times. While the arrival of new PET scanners (such as total body PET) is helping, there is always a need to improve reconstructed image quality due to the time and count limited imaging conditions. Artificial intelligence (AI) methods are now at the frontier of research for PET image reconstruction. While AI can learn the imaging physics as well as the noise in the data (when given sufficient examples), one of the most common uses of AI arises from exploiting databases of high-quality reference examples, to provide advanced noise compensation and resolution recovery. There are three main AI reconstruction approaches: (i) direct data-driven AI methods which rely on supervised learning from reference data, (ii) iterative (unrolled) methods which combine our physics and statistical models with AI learning from data, and (iii) methods which exploit AI with our known models, but crucially can offer benefits even in the absence of any example training data whatsoever. This article reviews these methods, considering opportunities and challenges of AI for PET reconstruction.

Comparison of post reconstruction- and reconstruction-based deep learning denoising methods in cardiac SPECT

Objective. The quality of myocardial perfusion SPECT (MPS) images is often hampered by low count statistics. Poor image quality might hinder reporting the studies and in the worst case lead to erroneous diagnosis. Deep learning (DL)-based methods can be used to improve the quality of the low count studies. DL can be applied in several different methods, which might affect the outcome. The aim of this study was to investigate the differences between post reconstruction- and reconstruction-based denoising methods.Approach. A UNET-type network was trained using ordered subsets expectation maximization (OSEM) reconstructed MPS studies acquired with half, quarter and eighth of full-activity. The trained network was applied as a post reconstruction denoiser (OSEM+DL) and it was incorporated into a regularized reconstruction algorithm as a deep learning penalty (DLP). OSEM+DL and DLP were compared against each other and against OSEM images without DL denoising in terms of noise level, myocardium-ventricle contrast and defect detection performance with signal-to-noise ratio of a non-prewhitening matched filter (NPWMF-SNR) applied to artificial perfusion defects inserted into defect-free clinical MPS scans. Comparisons were made using half-, quarter- and eighth-activity data.Main results. OSEM+DL provided lower noise level at all activities than other methods. DLP's noise level was also always lower than matching activity OSEM's. In addition, OSEM+DL and DLP outperformed OSEM in defect detection performance, but contrary to noise level ranking DLP had higher NPWMF-SNR overall than OSEM+DL. The myocardium-ventricle contrast was highest with DLP and lowest with OSEM+DL. Both OSEM+DL and DLP offered better image quality than OSEM, but visually perfusion defects were deeper in OSEM images at low activities.Significance. Both post reconstruction- and reconstruction-based DL denoising methods have great potential for MPS. The preference between these methods is a trade-off between smoother images and better defect detection performance.

Convex optimization algorithms in medical image reconstruction-in the age of AI

The past decade has seen the rapid growth of model based image reconstruction (MBIR) algorithms, which are often applications or adaptations of convex optimization algorithms from the optimization community. We review some state-of-the-art algorithms that have enjoyed wide popularity in medical image reconstruction, emphasize known connections between different algorithms, and discuss practical issues such as computation and memory cost. More recently, deep learning (DL) has forayed into medical imaging, where the latest development tries to exploit the synergy between DL and MBIR to elevate the MBIR's performance. We present existing approaches and emerging trends in DL-enhanced MBIR methods, with particular attention to the underlying role of convexity and convex algorithms on network architecture. We also discuss how convexity can be employed to improve the generalizability and representation power of DL networks in general.

Deep learning-based image reconstruction and post-processing methods in positron emission tomography for low-dose imaging and resolution enhancement

Image processing plays a crucial role in maximising diagnostic quality of positron emission tomography (PET) images. Recently, deep learning methods developed across many fields have shown tremendous potential when applied to medical image enhancement, resulting in a rich and rapidly advancing literature surrounding this subject. This review encapsulates methods for integrating deep learning into PET image reconstruction and post-processing for low-dose imaging and resolution enhancement. A brief introduction to conventional image processing techniques in PET is firstly presented. We then review methods which integrate deep learning into the image reconstruction framework as either deep learning-based regularisation or as a fully data-driven mapping from measured signal to images. Deep learning-based post-processing methods for low-dose imaging, temporal resolution enhancement and spatial resolution enhancement are also reviewed. Finally, the challenges associated with applying deep learning to enhance PET images in the clinical setting are discussed and future research directions to address these challenges are presented.