Increasing angular sampling through deep learning for stationary cardiac SPECT image reconstruction

Xie, Huidong; Thorn, Stephanie; Chen, X; Zhou, Bo; Liu, Hui; Liu, Zhao; Lee, Supum; Wang, Ge; Liu, Yi-Hwa; Sinusas, Albert J.; Liu, Chi

doi:10.1007/s12350-022-02972-z

Cited by 8 publications

(6 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, these scanners essentially represent limited-view imaging systems with truncated projections. Our previous work [15] proposed a multi-angle reconstruction approach to acquire multi-angle projections for improved image quality with detector gantry rotations. But obtaining multi-angle projections is not always feasible and complicates and extends the acquisition time.…”

Section: Discussionmentioning

confidence: 99%

“…Deep learning represents a new class of reconstruction algorithm [10] and may be an ideal tool to address the abovementioned limitations of anatomical-guided PVC methods. In the past years, convolutional-based neural networks have been implemented in various medical imaging applications such as low-dose CT [11], few-view CT [12], [13], low-dose SPECT [14], few-view SPECT [15], [16], attenuation map generations [17] etc. However, little attention has been given to deeplearning-based PVC.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Segmentation-free PVC for Cardiac SPECT using a Densely-connected Multi-dimensional Dynamic Network

Xie,

Liu,

Shi

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

In nuclear imaging, limited resolution causes partial volume effects (PVEs) that affect image sharpness and quantitative accuracy. Partial volume correction (PVC) methods incorporating high-resolution anatomical information from CT or MRI have been demonstrated to be effective. However, such anatomical-guided methods typically require tedious image registration and segmentation steps. Accurately segmented organ templates are also hard to obtain, particularly in cardiac SPECT imaging, due to the lack of hybrid SPECT/CT scanners with high-end CT and associated motion artifacts. Slight mis-registration/missegmentation would result in severe degradation in image quality after PVC. In this work, we develop a deeplearning-based method for fast cardiac SPECT PVC without anatomical information and associated organ segmentation. The proposed network involves a densely-connected multi-dimensional dynamic mechanism, allowing the convolutional kernels to be adapted based on the input images, even after the network is fully trained. Intramyocardial blood volume (IMBV) is introduced as an additional clinical-relevant loss function for network optimization. The proposed network demonstrated promising performance on 28 canine studies acquired on a GE Discovery NM/CT 570c dedicated cardiac SPECT scanner with a 64-slice CT using Technetium-99m-labeled red blood cells. This work showed that the proposed network with densely-connected dynamic mechanism produced superior results compared with the same network without such mechanism. Results also showed that the proposed network without anatomical information could produce images with statistically comparable IMBV measurements to the images generated by anatomical-guided PVC methods, which could be helpful in clinical translation.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Segmentation-free PVC for Cardiac SPECT using a Densely-connected Multi-dimensional Dynamic Network

Xie,

Liu,

Shi

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Porcine study reconstructed using one-angle data and four-angle data. Note that the image resolution significantly improved using four-angle data with the proposed multiangle reconstruction protocol in our previous work[4]. Rectangles enclosed by the L-shape arc represent CZT detector modules in the scanners.…”

mentioning

confidence: 88%

“…The proposed Transformer network was compared with the 3-D CNN network we proposed previously [4]. The 3-D CNN adapted a U-net-like [14] structure with four downsampling and four upsampling layers.…”

Section: U-net Structurementioning

confidence: 99%

“…The detector array is designed to acquire 19 projections simultaneously over a 180 • arch for stationary imaging. We previously developed an approach to acquire multiple projection angle sets by rotating and translating the detector and reconstructing the images with 76 projection views (19 projections × 4 angles) [4]. However, because rotating and translating the detector array are not practical in reality, a deep convolutional neural network (CNN) was also proposed to learn the mapping between one-angle image (19 projections) and four-angle images (76 projections) [4].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Deep-Learning-Based Few-Angle Cardiac SPECT Reconstruction Using Transformer

Xie

Thorn

Liu

et al. 2023

IEEE Trans. Radiat. Plasma Med. Sci.

View full text Add to dashboard Cite

Convolutional neural networks (CNNs) have been extremely successful in various medical imaging tasks. However, because the size of the convolutional kernel used in a CNN is much smaller than the image size, CNN has a strong spatial inductive bias and lacks a global understanding of the input images. Vision Transformer, a recently emerged network structure in computer vision, can potentially overcome the limitations of CNNs for image-reconstruction tasks. In this work, we proposed a slice-by-slice Transformer network (SSTrans-3D) to reconstruct cardiac single-photon emission computed tomography (SPECT) images from 3-D few-angle data. To be specific, the network reconstructs the whole 3-D volume using a sliceby-slice scheme. By doing so, SSTrans-3D alleviates the memory burden required by 3-D reconstructions using Transformer. The network can still obtain a global understanding of the image volume with the Transformer attention blocks. Finally, already reconstructed slices are used as the input to the network so that SSTrans-3D can potentially obtain more informative features from these slices. Validated on porcine, phantom, and human studies acquired using a GE dedicated cardiac SPECT scanner, the proposed method produced images with clearer heart cavity, higher cardiac defect contrast, and more accurate quantitative measurements on the testing data as compared with a deep U-net.

show abstract

Transformer-Based Dual-Domain Network for Few-View Dedicated Cardiac SPECT Image Reconstructions

Xie,

Zhou,

Chen

et al. 2023

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Increasing angular sampling through deep learning for stationary cardiac SPECT image reconstruction

Cited by 8 publications

References 18 publications

Segmentation-free PVC for Cardiac SPECT using a Densely-connected Multi-dimensional Dynamic Network

Segmentation-free PVC for Cardiac SPECT using a Densely-connected Multi-dimensional Dynamic Network

Deep-Learning-Based Few-Angle Cardiac SPECT Reconstruction Using Transformer

Transformer-Based Dual-Domain Network for Few-View Dedicated Cardiac SPECT Image Reconstructions

Contact Info

Product

Resources

About