生成对抗网络在医学图像转换领域的应用

常晓,蔡昕,杨光,聂生东

Applications of Generative Adversarial Networks in Medical Image Translation

Xiao CHANG,Xin CAI,Guang YANG,Sheng-dong NIE

图5 三项针对磁共振图像转换为CT图像的研究. (a)基于DCGAN，数据包括来自ADNI的大脑公开数据和自己获取的骨盆数据[35]；(b)分割任务与图像转换任务依靠GAN网络实现结合，数据也来自ADNI的大脑公开数据[36]；(c)仅MR引导的放射治疗[37]

Fig.5 Three researches on translating magnetic resonance image to CT image. (a) Based on DCGAN, and the datasets include the public brain dataset in ADNI database and their own pelvic dataset[35]; (b) Segmentation and image translation are all achieved by GAN, and the dataset is also the public brain dataset in the ADNI database[36]; (c) MR-only guided radiotherapy[37]