Evaluation of Image Processing Methods for Clinical Applications

Mimicking Clinical Data Using Conditional GANs
  • Hristina UzunovaEmail author
  • Sandra Schultz
  • Heinz Handels
  • Jan Ehrhardt
Conference paper
Part of the Informatik aktuell book series (INFORMAT)


While developing medical image applications, their accuracy is usually evaluated on a validation dataset, that generally differs from the real clinical data. Since clinical data does not contain ground truth annotations, it is impossible to approximate the real accuracy of the method. In this work, a cGAN-based method to generate realistically looking clinical data preserving the topology and thus ground truth of the validation set is presented. On the example of image registration of brain MRIs, we emphasize the necessity for the method and show that it enables evaluation of the accuracy on a clinical dataset. Furthermore, the topology preserving and realistic appearance of the generated images are evaluated and considered to be sufficient.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Goodfellow I, Pouget-Abadie J, Mirza M, et al. Generative adversarial nets. Adv Neural Inf Process Syst. 2014;27:2672–2680.Google Scholar
  2. 2.
    Armanious K, Yang C, Fischer M, et al. MedGAN: medical image translation using GANs. CoRR. 2018;abs/1806.06397.Google Scholar
  3. 3.
    Schlegl T, Seeböck P, Waldstein SM, et al. Unsupervised anomaly detection with generative adversarial networks to guide marker discovery. Proc IPMI. 2017; p. 146–157.Google Scholar
  4. 4.
    Zhu JY, Park T, Isola P, et al. Unpaired image-to-image translation using cycleconsistent adversarial networks. Proc ICCV. 2017; p. 2242–2251.Google Scholar
  5. 5.
    Isola P, Zhu JY, Zhou T, et al. Image-to-image translation with vonditional adversarial networks. Proc CVPR. 2017;00:5967–5976.Google Scholar
  6. 6.
    He K, Zhang X, Ren S, et al. Deep residual learning for image recognition. Proc CVPR. 2016; p. 770–778.Google Scholar
  7. 7.
    Shattuck DW, Mirza M, Adisetiyo V, et al. Construction of a 3D probabilistic atlas of human cortical structures. NeuroImage. 2008;39.Google Scholar
  8. 8.
    Menze BH, Jakab A, Bauer S, et al. The multimodal brain tumor image degmentation benchmark (BRATS). IEEE Trans Med Imaging. 2015;34(10):1993–2024.Google Scholar
  9. 9.
    Cohen JP, Luck M, Honari S. Distribution matching losses can hallucinate features in medical image Translation. Proc MICCAI. 2018;11070:529–536.Google Scholar
  10. 10.
    Ehrhardt J, Schmidt-Richberg A, Werner R, et al. Variational registration: a exible open-source ITK toolbox for nonrigid image registration. Proc BVM. 2015; p. 209–214.Google Scholar

Copyright information

© Springer Fachmedien Wiesbaden GmbH, ein Teil von Springer Nature 2019

Authors and Affiliations

  • Hristina Uzunova
    • 1
    Email author
  • Sandra Schultz
    • 1
  • Heinz Handels
    • 1
  • Jan Ehrhardt
    • 1
  1. 1.Institut für Medizinische InformatikUniversität zu LübeckLübeckDeutschland

Personalised recommendations