Meeting Banner
Abstract #0190

Quantitative susceptibility mapping using deep neural network

Jaeyeon Yoon1, Jingyu Ko1, Jingu Lee1, Hosan Jung1, Berkin Bilgic2, Kawin Setsompop2, and Jongho Lee1

1Department of Electrical and Computer Engineering, Seoul National University, Seoul, Republic of Korea, 2Department of Radiology, Harvard Medical School, Boston, MA, United States

In this study, we designed a deep neural network that functions as dipole deconvolution in QSM reconstruction. For label data, COSMOS reconstructed QSM maps were used so that the network produces ground truth like COSMOS results without streaking artifacts. The performance of our network was superior to conventional QSM results with lower RMSE for multiple head orientation input data.

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

Join Here