Multi-Scale Texture Loss for CT denoising with GANs

2024-03-25

arXiv.org

AI Summary

To address texture distortion and multi-scale detail loss in low-dose CT image denoising caused by conventional GANs, this paper proposes a differentiable multi-scale Gray-Level Co-occurrence Matrix (GLCM) texture loss function. We present the first end-to-end differentiable GLCM computation and integrate self-attention to enable multi-scale texture feature aggregation, thereby seamlessly embedding classical statistical texture modeling into GAN training. The proposed loss is validated across three distinct GAN architectures—DCGAN, StyleGAN2, and UNet-GAN—on one simulated and two real-world CT datasets. It consistently outperforms standard L1, VGG, and perceptual losses in terms of PSNR and SSIM, demonstrating both performance stability and architectural and dataset generalizability. The implementation is publicly available.

Technology Category

• Computer Vision: Generative Adversarial Networks (GANs) for Vision
• Machine Learning: Deep Generative Models & Autoencoders
• Natural Language Processing: Generation

Application Category

• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work
• Security and Privacy: Large-scale security measurements
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

Generative Adversarial Networks (GANs) have proved as a powerful framework for denoising applications in medical imaging. However, GAN-based denoising algorithms still suffer from limitations in capturing complex relationships within the images. In this regard, the loss function plays a crucial role in guiding the image generation process, encompassing how much a synthetic image differs from a real image. To grasp highly complex and non-linear textural relationships in the training process, this work presents a loss function that leverages the intrinsic multi-scale nature of the Gray-Level-Co-occurrence Matrix (GLCM). Although the recent advances in deep learning have demonstrated superior performance in classification and detection tasks, we hypothesize that its information content can be valuable when integrated into GANs' training. To this end, we propose a differentiable implementation of the GLCM suited for gradient-based optimization. Our approach also introduces a self-attention layer that dynamically aggregates the multi-scale texture information extracted from the images. We validate our approach by carrying out extensive experiments in the context of low-dose CT denoising, a challenging application that aims to enhance the quality of noisy CT scans. We utilize three publicly available datasets, including one simulated and two real datasets. The results are promising as compared to other well-established loss functions, being also consistent across three different GAN architectures. The code is available at: https://github.com/FrancescoDiFeola/DenoTextureLoss

Problem

• Generative Adversarial Networks
• Medical Image Processing
• Low-dose CT Denoising

Innovation

• Generative Adversarial Networks (GANs)
• Self-attention Mechanism
• Multi-scale Loss Function