Deep Retinal Image Understanding

By Kevis-Kokitsi Maninis, Jordi Pont-Tuset, Pablo Arbeláez, Luc Van Gool. International Conference on Medical Image Computing and Computer-Assisted Intervention, 2016

540 citations recorded by the source corpus.

This paper presents Deep Retinal Image Understanding (DRIU), a unified framework of retinal image analysis that provides both retinal vessel and optic disc segmentation. We make use of deep Convolutional Neural Networks (CNNs), which have proven revolutionary in other fields of computer vision such as object detection and image classification, and we bring their power to the study of eye fundus images. DRIU uses a base network architecture on which two set of specialized layers are trained to solve both the retinal vessel and optic disc segmentation. We present experimental validation, both qualitative and quantitative, in four public datasets for these tasks. In all of them, DRIU presents super-human performance, that is, it shows results more consistent with a gold standard than a second human annotator used as control.

Subfields: Ophthalmology, Radiology, Nuclear Medicine and Imaging, Signal Processing

View Paper: https://www.semanticscholar.org/paper/edcfcddee5e319dce8aa4057a784d5ba1969437c

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Ophthalmology

In Radiology, Nuclear Medicine and Imaging

In Signal Processing

Citations in the corpus

Standardized Assessment of Automatic Segmentation of White Matter Hyperintensities; Results of the WMH Segmentation Challenge (2019, background): “The VGG network had its fully connected layers replaced by a number of convolutional layers [34]–[37].”
A Three-Stage Deep Learning Model for Accurate Retinal Vessel Segmentation (2019, methodology): “Visually, the probability maps generated by the proposed model are much “cleaner” than those probabilitymaps in [30].”
Joint Optic Disc and Cup Segmentation Based on Multi-Label Deep Network and Polar Transformation (2018, background): “In fundus vessel segmentation, the deep learning systems [16]–[18] also achieve the state-of-the-art performances.”
Convolutional Oriented Boundaries: From Image Segmentation to High-Level Tasks (2016, background): “Our efforts on segmentation through CNNs have also found application in retinal image segmentation [17], obtaining state-of-the-art and super-human performance in vessel and optic disc segmentation, which further highlights their generality.”
Optic disc and cup segmentation methods for glaucoma detection with modification of U-Net convolutional neural network (2017, methodology): “Method from the paper [8] has several drawbacks.”
Video Object Segmentation without Temporal Information (2017, methodology): “Despite the use of skip connections [20], [40], [43], [73] to minimize the loss of spatial accuracy, we observe that OSVOS’ segmentations have”
Computer Methods and Programs in Biomedicine (unknown): “A uniﬁed framework of retinal image analysis that provides both retinal vessel and optic disc segmentation is proposed in [50] .”
End-to-End Adversarial Retinal Image Synthesis (2018, methodology): “9755 AUC on the DRIVE test set, a result aligned with state-of-the-art methods for retinal vessel segmentation [21]–[24].”
One-Shot Video Object Segmentation (2016, methodology): “Equation 1 allows training for imbalanced binary tasks [22, 53, 29, 30].”