GAN-based synthetic brain MR image generation
| Autor: | Changhee Han, Giancarlo Mauri, Wataru Shimoda, Leonardo Rundo, Hideki Nakayama, Ryosuke Araki, Hideaki Hayashi, Yujiro Furukawa, Shinichi Muramatsu |
|---|---|
| Přispěvatelé: | Han, C, Hayashi, H, Rundo, L, Araki, R, Shimoda, W, Muramatsu, S, Furukawa, Y, Mauri, G, Nakayama, H |
| Jazyk: | angličtina |
| Rok vydání: | 2018 |
| Předmět: |
Radiology
Nuclear Medicine and Imaging Generative Adversarial Networks Synthetic Medical Image Generation business.industry Computer science Biomedical Engineering Pattern recognition 02 engineering and technology Data Augmentation Visual Turing Test Generative Adversarial Network 030218 nuclear medicine & medical imaging 03 medical and health sciences 0302 clinical medicine Brain MRI 0202 electrical engineering electronic engineering information engineering Medical imaging 020201 artificial intelligence & image processing Brain magnetic resonance imaging Artificial intelligence Mr images business Focus (optics) Physician Training |
| Zdroj: | ISBI |
| Popis: | In medical imaging, it remains a challenging and valuable goal how to generate realistic medical images completely different from the original ones; the obtained synthetic images would improve diagnostic reliability, allowing for data augmentation in computer-assisted diagnosis as well as physician training. In this paper, we focus on generating synthetic multi-sequence brain Magnetic Resonance (MR) images using Generative Adversarial Networks (GANs). This involves difficulties mainly due to low contrast MR images, strong consistency in brain anatomy, and intra-sequence variability. Our novel realistic medical image generation approach shows that GANs can generate 128 χ 128 brain MR images avoiding artifacts. In our preliminary validation, even an expert physician was unable to accurately distinguish the synthetic images from the real samples in the Visual Turing Test. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|