Low-dose CT Denoising with Language-engaged Dual-space Alignment

TL;DR

This paper introduces LEDA, a novel LLM-guided dual-space alignment method that improves low-dose CT denoising by reducing over-smoothing and enhancing explainability through semantic and perceptual space alignment.

Contribution

The paper presents the first LLM-based scheme for low-dose CT denoising, leveraging language models to align denoised and normal dose images in perceptual and semantic spaces.

Findings

LEDA improves denoising performance on public datasets.

LEDA enhances explainability via language-level image understanding.

Experimental results show quantitative and qualitative improvements.

Abstract

While various deep learning methods were proposed for low-dose computed tomography (CT) denoising, they often suffer from over-smoothing, blurring, and lack of explainability. To alleviate these issues, we propose a plug-and-play Language-Engaged Dual-space Alignment loss (LEDA) to optimize low-dose CT denoising models. Our idea is to leverage large language models (LLMs) to align denoised CT and normal dose CT images in both the continuous perceptual space and discrete semantic space, which is the first LLM-based scheme for low-dose CT denoising. LEDA involves two steps: the first is to pretrain an LLM-guided CT autoencoder, which can encode a CT image into continuous high-level features and quantize them into a token space to produce semantic tokens derived from the LLM's vocabulary; and the second is to minimize the discrepancy between the denoised CT images and normal dose CT in terms of both encoded high-level features and quantized token embeddings derived by the LLM-guided CT autoencoder. Extensive experimental results on two public LDCT denoising datasets demonstrate that our LEDA can enhance existing denoising models in terms of quantitative metrics and qualitative evaluation, and also provide explainability through language-level image understanding. Source code is available at https://github.com/hao1635/LEDA.