Optimal Transport driven Cycle-consistent Generative Adversarial Network (OT-CycleGAN) for an accurate MR water-fat separation
Juan Meneses, Cristóbal Arrieta, Gabriel della Maggiora, Pablo Irarrázaval, Cristián Tejos N, Marcelo E. Andía, Sergio A. Uribe, Carlos Sing Long · Proceedings on CD-ROM - International Society for Magnetic Resonance in Medicine. Scientific Meeting and Exhibition/Proceedings of the International Society for Magnetic Resonance in Medicine, Scientific Meeting and Exhibition · 2023
MRI water-fat separation is an ill-posed inverse problem usually addressed using complex numerical methods. This problem is related to an MR signal physical model that includes the effects of R2* signal decay ratio and main magnetic field inhomogeneities (Δf), which induce non-linearities that are related to the ill-posedness of the inverse problem. In this work, we propose an Optimal Transport driven Cycle-consistent Generative Adversarial Networks (OT-CycleGAN) framework, which is physics-based, and could use partially labeled training data to estimate the non-linear components (R2* and Δf), and posteriorly compute the water-fat images through a conventional least-squares approach.