Abstract #0810

Generative Adversarial Network for T2-Weighted Fat Saturation MR Image Synthesis Using Bloch Equation-based Autoencoder Regularization

Sewon Kim¹, Hanbyol Jang², Seokjun Hong², Yeong Sang Hong³, Won C. Bae^4,5, Sungjun Kim*^3,6, and Dosik Hwang*²

¹Electrical and electronic engineering, Yonsei University, Seoul, Korea, Republic of, ²Yonsei University, Seoul, Korea, Republic of, ³Gangnam Severance Hospital, Seoul, Korea, Republic of, ⁴Department of Radiology, University of California-San Diego, San Diego, CA, United States, ⁵Department of Radiology, VA San Diego Healthcare System, San Diego, CA, United States, ⁶Yonsei University College of Medicine, Seoul, Korea, Republic of

We proposed a Bloch equation-based autoencoder regularization Generative Adversarial Network (BlochGAN) to generate T2-weighted fat saturation (T2 FS) images from T1-weighted (T1-w) and T2-weighted (T2-w) images for spine diagnosis. Our method can reduce the cost for acquiring multi-contrast images by reducing the number of contrasts to be scanned. BlochGAN properly generates the target contrast images by using GAN trained with the autoencoder regularization based on bloch equation, which is the basic principal of MR physics for identifying the physical basis of the contrasts. Our results demonstrate that BlochGAN achieved quantitatively and qualitatively superior performance compared to conventional methods.

This abstract and the presentation materials are available to members only; a login is required.

Join Here