AirNet: Fused analytical and iterative reconstruction with deep neural network regularization for sparse‐data CT

Abstract

Purpose: Sparse-data computed tomography (CT) frequently occurs, such as breast tomosynthesis, Carm CT, on-board four-dimensional cone-beam CT (4D CBCT), and industrial CT. However, sparse-data image reconstruction remains challenging due to highly undersampled data. This work develops a datadriven image reconstruction method for sparse-data CT using deep neural networks (DNN). Methods: The new method so-called AirNet is designed to incorporate the benefits from analytical reconstruction method (AR), iterative reconstruction method (IR), and DNN. It is built upon fused analytical and iterative reconstruction (AIR) that synergizes AR and IR via the optimization framework of modified proximal forward-backward splitting (PFBS). By unrolling PFBS into IR updates of CT data fidelity and DNN regularization with residual learning, AirNet utilizes AR such as FBP during the data fidelity, introduces dense connectivity into DNN regularization, and learns PFBS coefficients and DNN parameters that minimize the loss function during the training stage; and then AirNet with trained parameters can be used for end-to-end image reconstruction. Results: A CT atlas of 100 prostate scans was used to validate the AirNet in comparison with stateof-art DNN-based postprocessing and image reconstruction methods. The validation loss in AirNet had the fastest decreasing rate, owing to inherited fast convergence from AIR. AirNet was robust to noise in projection data and content differences between the training set and the images to be reconstructed. The impact of image quality on radiotherapy treatment planning was evaluated for both photon and proton therapy, and AirNet achieved the best treatment plan quality, especially for proton therapy. For example, with limited-angle data, the maximal target dose for AirNet was 109.5% in comparison with the ground truth 109.1%, while it was significantly elevated to 115.1% and 128.1% for FBPConvNet and LEARN, respectively.

Stabilizing deep tomographic reconstruction: Part A. Hybrid framework and experimental results

“…Second, the size of images reconstructed by an unrolled network is typically small. For example, the input image consists of small pixels for ADMM-net, 41 MetaInv-Net, 43 LEARN, 44 and AirNet, 46 limited by the memory size of the GPU. The reconstructed low-resolution results could not satisfy the requirement of many clinical applications, especially for CT imaging tasks.…”

Low-dose CT with deep learning regularization via proximal forward–backward splitting

“…In fact, the applied CNN is modified by concatenating all previous estimates of the latent image as the input. We replace CNN(x k+ 1 2 , θ 2 k ) with a densely-connected [34], which was utilized in our previous work for sparse-data CT [35] and shown to outperform the standard CNN. The problem of vanishing gradient can be addressed by the modification.…”

2-Step Sparse-View CT Reconstruction with a Domain-Specific Perceptual Network

“…Most OBL frameworks unroll IR, such as gradient descent, and learn the gradient of an image prior that is applied during each iteration. For SV-CT, Chen et al [4] demonstrated state-of-the-art performance with OBL. A more general approach to learning a data-driven inversion of linear inverse problems are Neumann networks introduced by Gilton et al [12].…”