eduzhai > Applied Sciences > Engineering >

Generating Fundus Fluorescence Angiography Images from Structure Fundus Images Using Generative Adversarial Networks

  • Save

... pages left unread,continue reading

Document pages: 16 pages

Abstract: Fluorescein angiography can provide a map of retinal vascular structure andfunction, which is commonly used in ophthalmology diagnosis, however, thisimaging modality may pose risks of harm to the patients. To help physiciansreduce the potential risks of diagnosis, an image translation method isadopted. In this work, we proposed a conditional generative adversarialnetwork(GAN) - based method to directly learn the mapping relationship betweenstructure fundus images and fundus fluorescence angiography images. Moreover,local saliency maps, which define each pixel s importance, are used to define anovel saliency loss in the GAN cost function. This facilitates more accuratelearning of small-vessel and fluorescein leakage features.

Please select stars to rate!


0 comments Sign in to leave a comment.

    Data loading, please wait...