Weight Encode Reconstruction Network for Computed Tomography in a Semi-Case-Wise and Learning-Based Way

TL;DR

This paper introduces WERNet, a semi-supervised learning-based method for CT reconstruction that accurately estimates voxel weights and demonstrates superior denoising and generalization capabilities compared to traditional methods.

Contribution

The study presents a novel semi-case-wise, self-supervised neural network architecture for CT reconstruction that effectively estimates voxel weights without labeled data.

Findings

Achieved cosine similarity > 0.999 with ground truth.

Demonstrated superior denoising compared to classic ART.

Encoder generalizes well to unseen cases.

Abstract

Classic algebraic reconstruction technology (ART) for computed tomography requires pre-determined weights of the voxels for projecting pixel values. However, such weight cannot be accurately obtained due to the limitation of the physical understanding and computation resources. In this study, we propose a semi-case-wise learning-based method named Weight Encode Reconstruction Network (WERNet) to tackle the issues mentioned above. The model is trained in a self-supervised manner without the label of a voxel set. It contains two branches, including the voxel weight encoder and the voxel attention part. Using gradient normalization, we are able to co-train the encoder and voxel set numerically stably. With WERNet, the reconstructed result was obtained with a cosine similarity greater than 0.999 with the ground truth. Moreover, the model shows the extraordinary capability of denoising comparing to the classic ART method. In the generalization test of the model, the encoder is transferable from a voxel set with complex structure to the unseen cases without the deduction of the accuracy.