Metal-conscious Embedding for CBCT Projection Inpainting

TL;DR

This paper introduces a hybrid neural network with metal-conscious embedding techniques to improve projection inpainting in CBCT images affected by metal artifacts, demonstrating enhanced accuracy on simulated and real data.

Contribution

It proposes a novel combination of Swin ViT and CNN with metal-conscious self-embedding methods for improved CBCT projection inpainting.

Findings

Achieved lowest mean absolute error of 0.079 in metal regions.

Attained highest peak signal-to-noise ratio of 42.346 in CBCT projections.

Enhanced inpainting performance on both simulated and real cadaver data.

Abstract

The existence of metallic implants in projection images for cone-beam computed tomography (CBCT) introduces undesired artifacts which degrade the quality of reconstructed images. In order to reduce metal artifacts, projection inpainting is an essential step in many metal artifact reduction algorithms. In this work, a hybrid network combining the shift window (Swin) vision transformer (ViT) and a convolutional neural network is proposed as a baseline network for the inpainting task. To incorporate metal information for the Swin ViT-based encoder, metal-conscious self-embedding and neighborhood-embedding methods are investigated. Both methods have improved the performance of the baseline network. Furthermore, by choosing appropriate window size, the model with neighborhood-embedding could achieve the lowest mean absolute error of 0.079 in metal regions and the highest peak signal-to-noise ratio of 42.346 in CBCT projections. At the end, the efficiency of metal-conscious embedding on both simulated and real cadaver CBCT data has been demonstrated, where the inpainting capability of the baseline network has been enhanced.