We propose a deep-learning based annotation efficient framework for vessel detection in ultra-widefield (UWF) fundus photography (FP) that does not require de novo labeled UWF FP vessel maps. Our approach utilizes concurrently captured UWF fluorescein angiography (FA) images, for which effective deep learning approaches have recently become available, and iterates between a multi-modal registration step and a weakly-supervised learning step. In the registration step, the UWF FA vessel maps detected with a pre-trained deep neural network (DNN) are registered with the UWF FP via parametric chamfer alignment. The warped vessel maps can be used as the tentative training data but inevitably contain incorrect (noisy) labels due to the differences between FA and FP modalities and the errors in the registration. In the learning step, a robust learning method is proposed to train DNNs with noisy labels. The detected FP vessel maps are used for the registration in the following iteration. The registration and the vessel detection benefit from each other and are progressively improved. Once trained, the UWF FP vessel detection DNN from the proposed approach allows FP vessel detection without requiring concurrently captured UWF FA images. We validate the proposed framework on a new UWF FP dataset, PRIMEFP20, and on existing narrow field FP datasets. Experimental evaluation, using both pixel wise metrics and the CAL metrics designed to provide better agreement with human assessment, shows that the proposed approach provides accurate vessel detection, without requiring manually labeled UWF FP training data.
The work was supported in part by a University of Rochester Research Award, by a distinguished researcher award from the New York state funded Rochester Center of Excellence in Data Science (CoE #3B C160189) at the University of Rochester, by an unrestricted grant to the Department of Ophthalmology from Research to Prevent Blindness, and grant P30EY001319- 35 from the National Institutes of Health.
ORCID of Submitting Author0000-0001-9735-9519
Submitting Author's InstitutionUniversity of Rochester
Submitting Author's CountryUnited States of America