From Local to Global: A Graph Framework for Retinal Artery/Vein Classification
- 23 June 2020
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Nanobioscience
- Vol. 19 (4), 589-597
- https://doi.org/10.1109/tnb.2020.3004481
Abstract
Fundus photography has been widely used for inspecting eye disorders by ophthalmologists or computer algorithms. Biomarkers related to retinal vessels plays an essential role to detect early diabetes. To quantify vascular biomarkers or the corresponding changes, an accurate artery and vein classification is necessary. In this work, we propose a new framework to boost local vessel classification with a global vascular network model using graph convolution. We compare our proposed method with two traditional state-of-the-art methods on a testing dataset of 750 images from the Maastricht Study. After incorporating global information, our model achieves the best accuracy of 86.45% compared to 85.5% from convolutional neural networks (CNN) and 82.9% from handcrafted pixel feature classification (HPFC). Our model also obtains the best area under receiver operating characteristic curve (AUC) of 0.95, compared to 0.93 from CNN and 0.90 from HPFC. The new classification framework has the advantage of easy deployment on top of local classification features. It corrects the local classification error by minimizing global classification error and it brings free additional classification performance.Keywords
Funding Information
- The Netherlands Organization for Scientific Research (629.001.003)
- Science and Technology Development Project of GuangZhou City (201604020016)
- Natural Science Foundation of China (61876194)
- Shanghai Key Lab of Digital Media Processing and Transmission (STCSM 18DZ2270700)
This publication has 36 references indexed in Scilit:
- Trainable COSFIRE filters for vessel delineation with application to retinal imagesMedical Image Analysis, 2015
- Complete Genome Sequence of a Newcastle Disease Virus Isolate from an Outbreak in Northern IndiaMicrobiology Resource Announcements, 2014
- Automated Method for Identification and Artery-Venous Classification of Vessel Trees in Retinal Vessel NetworksPLOS ONE, 2014
- Automated characterization of blood vessels as arteries and veins in retinal imagesComputerized Medical Imaging and Graphics, 2013
- An Automatic Graph-Based Approach for Artery/Vein Classification in Retinal ImagesIEEE Transactions on Image Processing, 2013
- Twisted Blood Vessels: Symptoms, Etiology and Biomechanical MechanismsJournal of Vascular Research, 2012
- Retinal vessel diameters and risk of strokeNeurology, 2006
- Luminosity and contrast normalization in retinal imagesMedical Image Analysis, 2005
- Gradient-based learning applied to document recognitionProceedings of the IEEE, 1998
- Fundus Photographic Risk Factors for Progression of Diabetic RetinopathyOphthalmology, 1991