Fetal MRI Synthesis via Balanced Auto-Encoder Based Generative Adversarial Networks.

  • Authors
  • Torrents-Barrena J, Piella G, Masoller N, Gratacos E, Eixarch E, Ceresa M, Ballester MAG.
  • UPF authors
  • PIELLA FENOY, GEMA; GONZALEZ BALLESTER, MIGUEL ANGEL;
  • Type
  • Scholarly articles
  • Journal títle
  • Conference Proceedings: Annual International Conference of the IEEE Engineering in Medicine and Biology Society
  • Publication year
  • 2018
  • Pages
  • 2599-2602
  • ISSN
  • 1557-170X
  • Publication State
  • Published
  • Abstract
  • Machine learning approaches for image analysis require large amounts of training imaging data. As an alternative, the use of realistic synthetic data reduces the high cost associated to medical image acquisition, as well as avoiding confidentiality and privacy issues, and consequently allows the creation of public data repositories for scientific purposes. Within the context of fetal imaging, we adopt an auto-encoder based Generative Adversarial Network for synthetic fetal MRI generation. The proposed architecture features a balanced power of the discriminator against the generator during training, provides an approximate convergence measure, and enables fast and robust training to generate high-quality fetal MRI in axial, sagittal and coronal planes. We demonstrate the feasibility of the proposed approach quantitatively and qualitatively by segmenting relevant fetal structures to assess the anatomical fidelity of the simulation, and performing a clinical verisimilitude study distinguishing the simulated data from the real images. The results obtained so far are promising, which makes further investigation on this new topic worthwhile.
  • Complete citation
  • Torrents-Barrena J, Piella G, Masoller N, Gratacos E, Eixarch E, Ceresa M, Ballester MAG.. Fetal MRI Synthesis via Balanced Auto-Encoder Based Generative Adversarial Networks.. Conference Proceedings: Annual International Conference of the IEEE Engineering in Medicine and Biology Society 2018; ( ): 2599-2602.
Bibliometric indicators
  • 0 times cited Scopus
  • Índex Scimago de 0.308(2018)