Swap-Net: A Memory-Efficient 2.5D Network for Sparse-View 3D Cone Beam CT Reconstruction

Abstract

Reconstructing 3D cone beam computed tomography (CBCT) images from a limited set of projections is an important inverse problem in many imaging applications from medicine to inertial confinement fusion (ICF). The performance of traditional methods such as filtered back projection (FBP) and model-based regularization is sub-optimal when the number of available projections is limited. In the past decade, deep learning (DL) has gained great popularity for solving CT inverse problems. A typical DL-based method for CBCT image reconstruction is to learn an end-to-end mapping by training a 2D or 3D network. However, 2D networks fail to fully use global information. While 3D networks are desirable, they become impractical as image sizes increase because of the high memory cost. This paper proposes Swap-Net, a memory-efficient 2.5D network for sparse-view 3D CBCT image reconstruction. Swap-Net uses a sequence of novel axes-swapping operations to produce 3D volume reconstruction in an end-to-end fashion without using full 3D convolutions. Simulation results show that Swap-Net consistently outperforms baseline methods both quantitatively and qualitatively in terms of reducing artifacts and preserving details of complex hydrodynamic simulations of relevance to the ICF community.

References in corpus (16)

• Deep Convolutional Neural Network for Inverse Problems in Imaging
• Low-Dose CT with a Residual Encoder-Decoder Convolutional Neural Network (RED-CNN)
• MoDL: Model Based Deep Learning Architecture for Inverse Problems
• Learned Primal-dual Reconstruction
• BM3D Frames and Variational Image Deblurring
• CNN-Based Projected Gradient Descent for Consistent Image Reconstruction
• Model based learning for accelerated, limited-view 3D photoacoustic tomography
• Image Reconstruction: From Sparsity to Data-adaptive Methods and Machine Learning
• SUNet: Swin Transformer UNet for Image Denoising
• A Deep Learning Reconstruction Framework for Differential Phase-Contrast Computed Tomography with Incomplete Data
• Limited View and Sparse Photoacoustic Tomography for Neuroimaging with Deep Learning
• PWLS-ULTRA: An Efficient Clustering and Learning-Based Approach for Low-Dose 3D CT Image Reconstruction
• A Cone-Beam X-Ray CT Data Collection designed for Machine Learning
• SGD-Net: Efficient Model-Based Deep Learning with Theoretical Guarantees
• An End-To-End-Trainable Iterative Network Architecture for Accelerated Radial Multi-Coil 2D Cine MR Image Reconstruction
• Local Models for Scatter Estimation and Descattering in Polyenergetic X-Ray Tomography