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, Marcelo E. Andía, Sergio Uribe, Carlos Sing LongProceedings 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 · DOI

Abstract

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 R 2 * 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 (R 2 * and Δf), and posteriorly compute the water-fat images through a conventional least-squares approach.
BibTeX
@inproceedings{meneses2023optimal,
  author = {Juan Meneses and Cristóbal Arrieta and Gabriel della Maggiora and Pablo Irarrázaval and Cristián Tejos and Marcelo E. Andía and Sergio Uribe and Carlos Sing Long},
  title = {{Optimal Transport driven Cycle-consistent Generative Adversarial Network (OT-CycleGAN) for an accurate MR water-fat separation}},
  booktitle = {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},
  year = {2023},
  doi = {10.58530/2022/2284},
}

Loading related content...

Figure