Unsupervised Deep Learning for Stroke Lesion Segmentation on Follow-up CT Based on Generative Adversarial Networks

Henk van Voorst, Praneeta R. Konduri, Laura M. van Poppel, Wouter van der Steen, P Matthijs van der Sluijs, E.m.h. Slot, Bart J. Emmer, Wim H. van Zwam, Yvo B.w.e.m. Roos, Charles B.l.m. Majoie, Greg Zaharchuk, Matthan W.a. Caan, Henk A. MarqueringAmerican 2022 · DOI

Abstract

An unsupervised generative adversarial network can be used to obtain automated infarct lesion segmentations with a moderate Dice similarity coefficient and good volumetric correspondence.
BibTeX
@inproceedings{voorst2022unsupervised,
  author = {Henk van Voorst and Praneeta R. Konduri and Laura M. van Poppel and Wouter van der Steen and P Matthijs van der Sluijs and E.M.H. Slot and Bart J. Emmer and Wim H. van Zwam and Yvo B.W.E.M. Roos and Charles B.L.M. Majoie and Greg Zaharchuk and Matthan W.A. Caan and Henk A. Marquering},
  title = {{Unsupervised Deep Learning for Stroke Lesion Segmentation on Follow-up CT Based on Generative Adversarial Networks}},
  booktitle = {American},
  year = {2022},
  doi = {10.3174/ajnr.a7582},
}

Loading related content...

Figure