QWD-GAN: Quality-aware Wavelet-driven GAN for Unsupervised Medical Microscopy Images Denoising

2025-09-19

AI Summary

Medical microscopic image denoising faces challenges including complex noise distributions, susceptibility to fine-detail loss, and insufficient algorithmic interpretability and clinical adaptability. To address these, we propose an unsupervised wavelet-driven generative adversarial network (WAVE-GAN). Our method introduces a multi-scale wavelet generator that adaptively models features across frequency bands, and a dual-branch discriminator integrated with difference-aware supervision to enhance preservation of high-frequency structures and textural details. Importantly, WAVE-GAN operates in a fully unpaired setting and maintains compatibility across diverse GAN architectures. Extensive experiments on multiple public medical microscopic image datasets demonstrate that WAVE-GAN consistently outperforms state-of-the-art methods, achieving superior performance in PSNR, SSIM, and visual fidelity. Notably, it excels at recovering clinically critical features—such as cell membranes and subcellular organelles—thereby improving diagnostic utility.

Technology Category

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

Application Category

• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

Image denoising plays a critical role in biomedical and microscopy imaging, especially when acquiring wide-field fluorescence-stained images. This task faces challenges in multiple fronts, including limitations in image acquisition conditions, complex noise types, algorithm adaptability, and clinical application demands. Although many deep learning-based denoising techniques have demonstrated promising results, further improvements are needed in preserving image details, enhancing algorithmic efficiency, and increasing clinical interpretability. We propose an unsupervised image denoising method based on a Generative Adversarial Network (GAN) architecture. The approach introduces a multi-scale adaptive generator based on the Wavelet Transform and a dual-branch discriminator that integrates difference perception feature maps with original features. Experimental results on multiple biomedical microscopy image datasets show that the proposed model achieves state-of-the-art denoising performance, particularly excelling in the preservation of high-frequency information. Furthermore, the dual-branch discriminator is seamlessly compatible with various GAN frameworks. The proposed quality-aware, wavelet-driven GAN denoising model is termed as QWD-GAN.

Problem

• Unsupervised denoising of medical microscopy images
• Preserving high-frequency details while removing noise
• Handling complex noise types in fluorescence-stained images

Innovation

• Wavelet-based multi-scale generator
• Dual-branch discriminator integration
• Unsupervised GAN denoising framework