SRU-Pix2Pix: A Fusion-Driven Generator Network for Medical Image Translation with Few-Shot Learning

2026-01-08

arXiv.org

AI Summary

This study addresses the limited clinical applicability of magnetic resonance imaging (MRI) due to prolonged acquisition times, high costs, and resolution constraints by proposing an enhanced Pix2Pix framework capable of high-quality medical image translation under extremely limited data regimes (<500 images). The approach innovatively integrates a Squeeze-and-Excitation Residual Network (SEResNet) with U-Net++ within the generator architecture and employs a streamlined PatchGAN discriminator, thereby strengthening channel-wise attention mechanisms and multi-scale feature fusion. Experimental results demonstrate that the proposed model significantly improves structural fidelity, image quality, and generalization across diverse MRI modality translation tasks, confirming its efficacy and novelty in data-scarce scenarios.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Multimodal Learning
• Natural Language Processing: Machine Translation, Multilinguality, Cross-Lingual NLP

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Web Mining and Content Analysis: Large pretrained models with web data
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Magnetic Resonance Imaging (MRI) provides detailed tissue information, but its clinical application is limited by long acquisition time, high cost, and restricted resolution. Image translation has recently gained attention as a strategy to address these limitations. Although Pix2Pix has been widely applied in medical image translation, its potential has not been fully explored. In this study, we propose an enhanced Pix2Pix framework that integrates Squeeze-and-Excitation Residual Networks (SEResNet) and U-Net++ to improve image generation quality and structural fidelity. SEResNet strengthens critical feature representation through channel attention, while U-Net++ enhances multi-scale feature fusion. A simplified PatchGAN discriminator further stabilizes training and refines local anatomical realism. Experimental results demonstrate that under few-shot conditions with fewer than 500 images, the proposed method achieves consistent structural fidelity and superior image quality across multiple intra-modality MRI translation tasks, showing strong generalization ability. These results suggest an effective extension of Pix2Pix for medical image translation.

Problem

• medical image translation
• few-shot learning
• structural fidelity
• image quality
• MRI

Innovation

• SEResNet
• U-Net++
• Few-shot learning
• Medical image translation
• Channel attention