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

Abstract

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.

Deep learning-based PET image denoising and reconstruction: a review

“…Gong et al proposed MAPEM-Net, which can be easily implemented by incorporating a potential function into neural-network optimization [203]. In addition, various other iterative PET image reconstruction algorithms have been proposed for PET and SPECT [204][205][206][207][208][209][210][211][212][213][214][215][216].…”

A Novel Implicit Neural Representation for Volume Data

“…The Ref. [32] suggests using vision transformers to remove restrictions of CNNs in image reconstruction. The slice-by-slice transformer network (SSTrans-3D) is a transformer-based technique that reconstructs 3D single-photon emission computed tomography (SPECT) images.…”