Show simple item record

dc.contributor.authorLiu, Yi Chieh
dc.contributor.authorYang, Hao Hsiang
dc.contributor.authorHuck Yang, C. H.
dc.contributor.authorHuang, Jia-Hong
dc.contributor.authorTian, Meng
dc.contributor.authorMorikawa, Hiromasa
dc.contributor.authorTsai, Yi Chang James
dc.contributor.authorTegner, Jesper
dc.identifier.citationLiu, Y.-C., Yang, H.-H., Huck Yang, C.-H., Huang, J.-H., Tian, M., Morikawa, H., … Tegnèr, J. (2019). Synthesizing New Retinal Symptom Images by Multiple Generative Models. Lecture Notes in Computer Science, 235–250. doi:10.1007/978-3-030-21074-8_19
dc.description.abstractAge-Related Macular Degeneration (AMD) is an asymptomatic retinal disease which may result in loss of vision. There is limited access to high-quality relevant retinal images and poor understanding of the features defining sub-classes of this disease. Motivated by recent advances in machine learning we specifically explore the potential of generative modeling, using Generative Adversarial Networks (GANs) and style transferring, to facilitate clinical diagnosis and disease understanding by feature extraction. We design an analytic pipeline which first generates synthetic retinal images from clinical images; a subsequent verification step is applied. In the synthesizing step we merge GANs (DCGANs and WGANs architectures) and style transferring for the image generation, whereas the verified step controls the accuracy of the generated images. We find that the generated images contain sufficient pathological details to facilitate ophthalmologists’ task of disease classification and in discovery of disease relevant features. In particular, our system predicts the drusen and geographic atrophy sub-classes of AMD. Furthermore, the performance using CFP images for GANs outperforms the classification based on using only the original clinical dataset. Our results are evaluated using existing classifier of retinal diseases and class activated maps, supporting the predictive power of the synthetic images and their utility for feature extraction. Our code examples are available online. (
dc.publisherSpringer Nature
dc.rightsThe final publication is available at Springer via 10.1007/978-3-030-21074-8_19
dc.titleSynthesizing New Retinal Symptom Images by Multiple Generative Models
dc.typeConference Paper
dc.contributor.departmentBiological and Environmental Sciences and Engineering (BESE) Division
dc.contributor.departmentBioscience Program
dc.contributor.departmentEarth Science and Engineering
dc.contributor.departmentEarth Science and Engineering Program
dc.conference.date2018-12-02 to 2018-12-06
dc.conference.name14th Asian Conference on Computer Vision, ACCV 2018
dc.conference.locationPerth, WA, AUS
dc.contributor.institutionNational Taiwan University, Taipei, Taiwan
dc.contributor.institutionGeorgia Institute of Technology, Atlanta, GA, USA
dc.contributor.institutionDepartment of Ophthalmology, Bern University Hospital, Bern, Switzerland
dc.contributor.institutionUnit of Computational Medicine, Center for Molecular Medicine, Department of Medicine, Karolinska Institutet, Stockholm, Sweden
kaust.personHuck Yang, C. H.
kaust.personHuang, Jia-Hong
kaust.personMorikawa, Hiromasa
kaust.personTegner, Jesper
display.relations<b>Is Supplemented By:</b><br/> <ul><li><i>[Software]</i> <br/> Title: huckiyang/EyeNet-GANs: ACCV'18 workshop - Synthesizing New Retinal Symptom Images by Multiple Generative Models. Publication Date: 2018-12-03. github: <a href="" >huckiyang/EyeNet-GANs</a> Handle: <a href="" >10754/668079</a></a></li></ul>

Files in this item


This item appears in the following Collection(s)

Show simple item record

The final publication is available at Springer via 10.1007/978-3-030-21074-8_19
Except where otherwise noted, this item's license is described as The final publication is available at Springer via 10.1007/978-3-030-21074-8_19