Deep Kernel Representation for Image Reconstruction in PET

Li, Siqi; Wang, Guobao

doi:10.1109/tmi.2022.3176002

Cited by 16 publications

(10 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, the construction of a spatial kernel itself can also be modified by using a different kernel function, e.g., using a pre-defined wavelet representation [40] or a neural-network representation [43]. The kernel construction can be further trained using deep learning as demonstrated in our recent work [44]. It is worth noting that all these methods are aimed at improving K and are therefore complementary to the proposed neural KEM which improves α in (5).…”

Section: Discussionmentioning

confidence: 99%

Neural KEM: A Kernel Method With Deep Coefficient Prior for PET Image Reconstruction

Gong

Badawi

et al. 2023

IEEE Trans. Med. Imaging

View full text Add to dashboard Cite

Image reconstruction of low-count positron emission tomography (PET) data is challenging. Kernel methods address the challenge by incorporating image prior information in the forward model of iterative PET image reconstruction. The kernelized expectationmaximization (KEM) algorithm has been developed and demonstrated to be effective and easy to implement. A common approach for a further improvement of the kernel method would be adding an explicit regularization, which however leads to a complex optimization problem. In this paper, we propose an implicit regularization for the kernel method by using a deep coefficient prior, which represents the kernel coefficient image in the PET forward model using a convolutional neural-network. To solve the maximumlikelihood neural network-based reconstruction problem, we apply the principle of optimization transfer to derive a neural KEM algorithm. Each iteration of the algorithm consists of two separate steps: a KEM step for image update from the projection data and a deep-learning step in the image domain for updating the kernel coefficient image using the neural network. This optimization algorithm is guaranteed to monotonically increase the data likelihood. The results from computer simulations and real patient data have demonstrated that the neural KEM can outperform existing KEM and deep image prior methods. I. INTRODUCTION T OMOGRAPHIC image reconstruction for positron emission tomography (PET) is challenging because of the ill-conditioned problem and low counting statistics [1]. The kernel method addresses this challenge by integrating image prior information into the forward model of PET image

show abstract

Section: Discussionmentioning

confidence: 99%

Neural KEM: A Kernel Method With Deep Coefficient Prior for PET Image Reconstruction

Gong

Badawi

et al. 2023

IEEE Trans. Med. Imaging

View full text Add to dashboard Cite

show abstract

“…A final important category, related to the example of using the kernel method with the DIP as previously shown in Figure 10 (and Table 4 ), is using a deep network to learn the best features to use for forming the kernel matrix used in KEM, called the deep kernel representation. 78 Training data are derived from the single dataset to reconstruct (by generating low-count versions of the data), and the method elegantly learns what are the most appropriate features to pick out from an input prior image such that when these features are used in a voxel-level similarity comparison to form spatial basis functions for KEM, the output image best matches the full-count high quality reference. These basis functions are then used with standard KEM for reconstructing the whole dataset.…”

Section: Ai Methods Without Training Datamentioning

confidence: 99%

AI for PET image reconstruction

Reader

Pan

2023

The British Journal of Radiology

View full text Add to dashboard Cite

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.

show abstract

“…In deep learning efforts, there are two groups of approaches: direct and indirect methods. Direct deep learning methods, such as DeepPET, 7 DPIR-Net 8 and Li, 9 in PET image reconstruction have the potential to achieve high-quality results and reduce computation time, but they are limited by the need of extensive training data, generalization issues, and lack of interpretability. 10 Indirect deep learning methods, such as FBPNet, 11 Chen 12 and Xie, 13 involve using surrogate tasks or incorporating prior information to address challenges like limited data availability, but they are susceptible to the accuracy of prior information.…”

Section: Introductionmentioning

confidence: 99%

Dynamic PET image reconstruction using deep physiology prior

Chen,

Hu,

Liu

2024

Medical Imaging 2024: Physics of Medical Imaging

View full text Add to dashboard Cite

Dynamic positron emission tomography (PET) imaging can provide information about metabolic changes over time, and is widely used in clinical diagnosis and cancer treatment. However, the existing deep learning methods for PET image reconstruction mainly focus on the static mapping paradigm between sinogram and radioactivity concentration distribution, which ignores the inherent dynamic activation process of tracers. In this paper, we establish a physiological model based deep learning framework for dynamic PET image reconstruction using deep physiology prior. First, the objective functions of our physiological model are combined with static mapping and the dynamic activation process of tracers. Then, a data-driven Adaptive Kalman Inspired Network (AKIN) is adopted to solve the proposed objective functions. Specifically, the AKIN consists of three components: a Prediction Net is employed for directly predicting the prior estimation; a Projection Net is employed for predicting the current estimation of the observations based on the prior estimation, furthermore, a Kalman Gain Net (KNet) is employed for adaptively learning the gain coefficient. The experiment of simulation data demonstrates that the proposed method has substantial noise reduction in temporal and spatial domains, outperforming other methods like maximum likelihood expectation maximization, kernel expectation maximization method and DeepPET.

show abstract

Deep Kernel Representation for Image Reconstruction in PET

Cited by 16 publications

References 37 publications

Neural KEM: A Kernel Method With Deep Coefficient Prior for PET Image Reconstruction

Neural KEM: A Kernel Method With Deep Coefficient Prior for PET Image Reconstruction

AI for PET image reconstruction

Dynamic PET image reconstruction using deep physiology prior

Contact Info

Product

Resources

About