GLFC: Unified Global-Local Feature and Contrast Learning with Mamba-Enhanced UNet for Synthetic CT Generation from CBCT

TL;DR

This paper introduces a novel framework called GLFC that combines a Mamba-Enhanced UNet and a Multiple Contrast Loss to improve synthetic CT generation from CBCT by effectively capturing global and local features and contrasts.

Contribution

The paper proposes a new GLFC framework with a Mamba-Enhanced UNet and a Multiple Contrast Loss for superior sCT generation from CBCT images.

Findings

Significant SSIM improvement from 77.91% to 91.50%.

Outperforms existing methods in synthetic CT quality.

Effective global and local feature learning demonstrated.

Abstract

Generating synthetic Computed Tomography (CT) images from Cone Beam Computed Tomography (CBCT) is desirable for improving the image quality of CBCT. Existing synthetic CT (sCT) generation methods using Convolutional Neural Networks (CNN) and Transformers often face difficulties in effectively capturing both global and local features and contrasts for high-quality sCT generation. In this work, we propose a Global-Local Feature and Contrast learning (GLFC) framework for sCT generation. First, a Mamba-Enhanced UNet (MEUNet) is introduced by integrating Mamba blocks into the skip connections of a high-resolution UNet for effective global and local feature learning. Second, we propose a Multiple Contrast Loss (MCL) that calculates synthetic loss at different intensity windows to improve quality for both soft tissues and bone regions. Experiments on the SynthRAD2023 dataset demonstrate that GLFC improved the SSIM of sCT from 77.91% to 91.50% compared with the original CBCT, and significantly outperformed several existing methods for sCT generation. The code is available at https://github.com/HiLab-git/GLFC