De-GAN - Dynamic Parameter Tuned GAN for 3D Medical Image Segmentation: A Step Towards Generalisation

2026-09-15

AI Summary

本文提出DE-GAN，通过动态卷积、特征混合和坐标编码生成适应性FLAIR图像，以解决因对比度低及域偏移导致的脑肿瘤分割难题。

Abstract

Brain tumor segmentation remains difficult because enhancing tumor (ET) has low contrast and overlaps surrounding tissue, while scanner and site variation causes domain shift. We propose DE-GAN, a contrast-enhancing conditional GAN that combines input-adaptive dynamic convolutions, style-aware feature mixing, and coordinate encoding to synthesize slice-adaptive FLAIR images. A label-guided, class-conditional target separates tumor-core (TC) and ET intensities while preserving anatomy. The generated FLAIR is concatenated with the original MR modalities and used to train a 3D U-Net. Across BraTS 2015, 2018, and 2019, DE-GAN improves segmentation over the baseline and static EnhGAN replacement on most reported TC/ET metrics, with the largest gains from retaining both original and enhanced FLAIR. Code and pretrained models are available at https://github.com/zkhansuri-ui/DE-GAN.

Problem

• brain tumor segmentation
• contrast enhancement
• domain shift

Innovation

• Dynamic Convolution
• Style-aware Feature Mixing
• Coordinate Encoding
• FLAIR Image Synthesis