Dual Path Learning -- learning from noise and context for medical image denoising

TL;DR

This paper introduces a Dual-Pathway Learning model for medical image denoising that effectively combines noise characteristics and contextual information, demonstrating improved robustness across multiple modalities and noise types.

Contribution

It proposes a novel dual-pathway architecture that integrates noise and context for enhanced medical image denoising, extending beyond single-modality and noise-specific methods.

Findings

DPL improves PSNR by 3.35% over UNet on Gaussian noise.

DPL is effective across multiple imaging modalities.

The model demonstrates robustness and generalizability.

Abstract

Medical imaging plays a critical role in modern healthcare, enabling clinicians to accurately diagnose diseases and develop effective treatment plans. However, noise, often introduced by imaging devices, can degrade image quality, leading to misinterpretation and compromised clinical outcomes. Existing denoising approaches typically rely either on noise characteristics or on contextual information from the image. Moreover, they are commonly developed and evaluated for a single imaging modality and noise type. Motivated by Geng et.al CNCL, which integrates both noise and context, this study introduces a Dual-Pathway Learning (DPL) model architecture that effectively denoises medical images by leveraging both sources of information and fusing them to generate the final output. DPL is evaluated across multiple imaging modalities and various types of noise, demonstrating its robustness and generalizability. DPL improves PSNR by 3.35% compared to the baseline UNet when evaluated on Gaussian noise and trained across all modalities. The code is available at 10.5281/zenodo.15836053.