We proposed a model-based generative adversarial network for quantitative susceptibility mapping. Total 30 scans from six healthy subjects, acquired at five different head orientations, were employed for network training. The trained network provided superior image quality and accuracy quantification compared to recently developed QSM reconstruction methods. The proposed method showed excellent tissue susceptibility contrast and artifact suppression on the QSM images of patients with hemorrhage and multiple sclerosis, demonstrating potential clinical application in the future.
This abstract and the presentation materials are available to members only; a login is required.