Paper
24 March 2023 Cross-modal image segmentation and synthesis in medical imaging
Xiaolong Wu, Zhimin Gan, Jiaxi Xie, Haiyi Lou, Zhengxing Yan, Qiqi Jia, Yingchun Yuan
Author Affiliations +
Proceedings Volume 12611, Second International Conference on Biological Engineering and Medical Science (ICBioMed 2022); 126110R (2023) https://doi.org/10.1117/12.2669536
Event: International Conference on Biological Engineering and Medical Science (ICBioMed2022), 2022, Oxford, United Kingdom
Abstract
Medical image is an important type of evidence for diagnosising disease. However, some patients can not receive a complete radiological examination due to the damage of radiation, medical cost or individual differences, so the transmembrane of medical images is essential. Genrative adversarial network (GAN) is an unsupervised deep learning model, which is widely used in the synthesis of medical images. In this research, paired data set which has MRI images of brain is used to compare the ability of transforming T1 of T2 images by pix2pix and cycleGAN. Peak signal to ratio (PSNR) and Mean absolute error (MAE) are used to evaluate the quantity of predicted T2 images. The weight of idenitity loss and hyperpremeters are adjusted to explore a better model of cycleGAN. The conclusion is that cycleGAN with low weight of identity loss is better than that with high weight of identity loss.
© (2023) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Xiaolong Wu, Zhimin Gan, Jiaxi Xie, Haiyi Lou, Zhengxing Yan, Qiqi Jia, and Yingchun Yuan "Cross-modal image segmentation and synthesis in medical imaging", Proc. SPIE 12611, Second International Conference on Biological Engineering and Medical Science (ICBioMed 2022), 126110R (24 March 2023); https://doi.org/10.1117/12.2669536
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Gallium nitride

Magnetic resonance imaging

Medical imaging

Data modeling

Education and training

Statistical modeling

Adversarial training

Back to Top