数据与计算发展前沿2024，Vol.6Issue(2)：80-88,9.DOI:10.11871/jfdc.issn.2096-742X.2024.02.008
基于域无关循环生成对抗网络的跨模态医学影像生成
Domain Independent Cycle-GAN for Cross Modal Medical Image Generation
摘要
Abstract
[Objective]To solve the problem of misaligned and inaccurate image structure generated by training with unpaired data in cross-modal medical image generation tasks,[Methods]this arti-cle proposes a cross-modal medical image generation model based on domain independent cy-clic generation adversarial network,which constrains the structural consistency of images be-fore and after modal transformation by aligning the intermediate features during cyclic genera-tion.[Results]The experimental results on the brain CT-MRI dataset show that the method pro-posed in this article can improve the consistency of image structure before and after cross-modal transformation,thereby improving the quality of cross-modal medical image generation.[Limitations]This study has conducted a large number of experiments on the multimodal brain dataset.Further verification of its universality is needed in other datasets.[Conclusions]The method proposed in this article outperforms the current best performing cross-modal medical image generation model in various metrics for measuring the quality of generated images.关键词
生成式对抗网络/医学影像生成/自注意力机制Key words
generative adversarial networks/medical image synthesis/self-attention mechanism引用本文复制引用
李浩鹏,周琬婷,陈玉,张曼..基于域无关循环生成对抗网络的跨模态医学影像生成[J].数据与计算发展前沿,2024,6(2):80-88,9.基金项目
国家自然科学基金面上项目"面向脑动静脉畸形辅助诊疗的多模态医学影像分析"(62376037) （62376037）
国家自然科学基金青年项目"移动场景眼部多模态生物特征识别"(62006227) （62006227）
辽宁省感知与理解人工智能重点实验室开放课题基金"复杂医学影像分析与生成研究"(20230006) （20230006）
