ROSE: A Retinal OCT-Angiography Vessel Segmentation Dataset and New Model

Optical Coherence Tomography Angiography (OCTA) is a non-invasive imaging technique that has been increasingly used to image the retinal vasculature at capillary level resolution. However, automated segmentation of retinal vessels in OCTA has been under-studied due to various challenges such as low capillary visibility and high vessel complexity, despite its significance in understanding many vision-related diseases. In addition, there is no publicly available OCTA dataset with manually graded vessels for training and validation of segmentation algorithms. To address these issues, for the first time in the field of retinal image analysis we construct a dedicated Retinal OCTA SEgmentation dataset (ROSE), which consists of 229 OCTA images with vessel annotations at either centerline-level or pixel level. This dataset with the source code has been released for public access to assist researchers in the community in undertaking research in related topics. Secondly, we introduce a novel split-based coarse-to-fine vessel segmentation network for OCTA images (OCTA-Net), with the ability to detect thick and thin vessels separately. In the OCTA-Net, a split-based coarse segmentation module is first utilized to produce a preliminary confidence map of vessels, and a split-based refined segmentation module is then used to optimize the shape/contour of the retinal microvasculature. We perform a thorough evaluation of the state-of-the-art vessel segmentation models and our OCTA-Net on the constructed ROSE dataset. The experimental results demonstrate that our OCTA-Net yields better vessel segmentation performance in OCTA than both traditional and other deep learning methods. In addition, we provide a fractal dimension analysis on the segmented microvasculature, and the statistical analysis demonstrates significant differences between the healthy control and Alzheimer's Disease group. This consolidates that the analysis of retinal microvasculature may offer a new scheme to study various neurodegenerative diseases.

[1]  Aaron Y. Lee,et al.  Associations between recent and established ophthalmic conditions and risk of Alzheimer's disease , 2018, Alzheimer's & Dementia.

[2]  Gabor Mark Somfai,et al.  Retinal microvascular network alterations: potential biomarkers of cerebrovascular and neural diseases. , 2017, American journal of physiology. Heart and circulatory physiology.

[3]  M. Schwartz,et al.  The retina as a window to the brain—from eye research to CNS disorders , 2013, Nature Reviews Neurology.

[4]  Ke Chen,et al.  Automated Vessel Segmentation Using Infinite Perimeter Active Contour Model with Hybrid Region Information with Application to Retinal Images , 2015, IEEE Transactions on Medical Imaging.

[5]  Mona Sharifi Sarabi,et al.  An Automated 3D Analysis Framework for Optical Coherence Tomography Angiography , 2019 .

[6]  Ling Shao,et al.  ET-Net: A Generic Edge-aTtention Guidance Network for Medical Image Segmentation , 2019, MICCAI.

[7]  Yongtian Wang,et al.  Intensity and Compactness Enabled Saliency Estimation for Leakage Detection in Diabetic and Malarial Retinopathy , 2017, IEEE Transactions on Medical Imaging.

[8]  Stephen Lin,et al.  DeepVessel: Retinal Vessel Segmentation via Deep Learning and Conditional Random Field , 2016, MICCAI.

[9]  Kenneth Falconer,et al.  Fractal Geometry: Mathematical Foundations and Applications , 1990 .

[10]  Ting Liu,et al.  Segmentation and quantification of blood vessels for OCT-based micro-angiograms using hybrid shape/intensity compounding. , 2015, Microvascular research.

[11]  Yong Liu,et al.  Extreme learning machine with hybrid cost function of G-mean and probability for imbalance learning , 2020, International Journal of Machine Learning and Cybernetics.

[12]  Josien P. W. Pluim,et al.  Robust Retinal Vessel Segmentation via Locally Adaptive Derivative Frames in Orientation Scores , 2016, IEEE Transactions on Medical Imaging.

[13]  Shihao Zhang,et al.  Attention Guided Network for Retinal Image Segmentation , 2019, MICCAI.

[14]  T. MacGillivray,et al.  Fractal analysis of the retinal vascular network in fundus images , 2007, 2007 29th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[15]  D. Grewal,et al.  Assessment of Differences in Retinal Microvasculature Using OCT Angiography in Alzheimer's Disease: A Twin Discordance Report. , 2018, Ophthalmic surgery, lasers & imaging retina.

[16]  Mona Sharifi Sarabi,et al.  3D Shape Modeling and Analysis of Retinal Microvasculature in OCT-Angiography Images , 2020, IEEE Transactions on Medical Imaging.

[17]  Jiong Zhang,et al.  3D Surface-Based Geometric and Topological Quantification of Retinal Microvasculature in OCT-Angiography via Reeb Analysis , 2019, MICCAI.

[18]  George Azzopardi,et al.  Trainable COSFIRE filters for vessel delineation with application to retinal images , 2015, Medical Image Anal..

[19]  Szilárd Kiss,et al.  Comparison of ultra-widefield fluorescein angiography with the Heidelberg Spectralis® noncontact ultra-widefield module versus the Optos® Optomap® , 2013, Clinical ophthalmology.

[20]  Gözde B. Ünal,et al.  A Higher-Order Tensor Vessel Tractography for Segmentation of Vascular Structures , 2015, IEEE Transactions on Medical Imaging.

[21]  Jeong W. Pak,et al.  Prevalence and Severity of Artifacts in Optical Coherence Tomographic Angiograms. , 2019, JAMA ophthalmology.

[22]  Jennifer I. Lim,et al.  Color Fundus Image Guided Artery-Vein Differentiation in Optical Coherence Tomography Angiography , 2018, Investigative ophthalmology & visual science.

[23]  Mehmet Bulut,et al.  Evaluation of optical coherence tomography angiographic findings in Alzheimer’s type dementia , 2017, British Journal of Ophthalmology.

[24]  Qingjie Liu,et al.  Road Extraction by Deep Residual U-Net , 2017, IEEE Geoscience and Remote Sensing Letters.

[25]  Fred K Chen,et al.  Classification of image artefacts in optical coherence tomography angiography of the choroid in macular diseases , 2016, Clinical & experimental ophthalmology.

[26]  Ayman El-Baz,et al.  Automatic blood vessels segmentation based on different retinal maps from OCTA scans , 2017, Comput. Biol. Medicine.

[27]  Wang Xiancheng,et al.  Retina blood vessel segmentation using a U-net based Convolutional neural network , 2018 .

[28]  T. Wong,et al.  Quantitative Assessment of Early Diabetic Retinopathy Using Fractal Analysis , 2009, Diabetes Care.

[29]  Qiang Chen,et al.  Image Projection Network: 3D to 2D Image Segmentation in OCTA Images , 2020, IEEE Transactions on Medical Imaging.

[30]  Shenghua Gao,et al.  CE-Net: Context Encoder Network for 2D Medical Image Segmentation , 2019, IEEE Transactions on Medical Imaging.

[31]  Erik J. Bekkers,et al.  Retinal vessel delineation using a brain-inspired wavelet transform and random forest , 2017, Pattern Recognit..

[32]  Sophie Lemmens,et al.  Systematic Review on Fractal Dimension of the Retinal Vasculature in Neurodegeneration and Stroke: Assessment of a Potential Biomarker , 2020, Frontiers in Neuroscience.

[33]  Li Chen,et al.  Dense Dilated Network With Probability Regularized Walk for Vessel Detection , 2020, IEEE Transactions on Medical Imaging.

[34]  David Huang,et al.  Enhanced Quantification of Retinal Perfusion by Improved Discrimination of Blood Flow From Bulk Motion Signal in OCTA , 2018, Translational vision science & technology.

[35]  J. Fujimoto,et al.  IMAGE ARTIFACTS IN OPTICAL COHERENCE TOMOGRAPHY ANGIOGRAPHY , 2015, Retina.

[36]  R. A. Leitgeb,et al.  En face optical coherence tomography: a technology review [Invited]. , 2019, Biomedical optics express.

[37]  D. Sarraf,et al.  Retinal Capillary Density and Foveal Avascular Zone Area Are Age-Dependent: Quantitative Analysis Using Optical Coherence Tomography Angiography. , 2016, Investigative ophthalmology & visual science.

[38]  Sheng Tang,et al.  Global-residual and Local-boundary Refinement Networks for Rectifying Scene Parsing Predictions , 2017, IJCAI.

[39]  Tuan D. Pham,et al.  DUNet: A deformable network for retinal vessel segmentation , 2018, Knowl. Based Syst..

[40]  Ruikang K. Wang,et al.  Minimizing projection artifacts for accurate presentation of choroidal neovascularization in OCT micro-angiography. , 2015, Biomedical optics express.

[41]  Kevin W Eliceiri,et al.  NIH Image to ImageJ: 25 years of image analysis , 2012, Nature Methods.

[42]  Jie Wang,et al.  Maximum value projection produces better en face OCT angiograms than mean value projection. , 2018, Biomedical optics express.

[43]  Maqsood Yaqub,et al.  Optical coherence tomography angiography in preclinical Alzheimer’s disease , 2019, British Journal of Ophthalmology.

[44]  Alejandro F. Frangi,et al.  CS-Net: Channel and Spatial Attention Network for Curvilinear Structure Segmentation , 2019, MICCAI.

[45]  Vijayan K. Asari,et al.  Recurrent Residual Convolutional Neural Network based on U-Net (R2U-Net) for Medical Image Segmentation , 2018, ArXiv.

[46]  Chongruo Wu,et al.  ResNeSt: Split-Attention Networks , 2020, 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

[47]  Edmund Arthur,et al.  Investigating Multimodal Diagnostic Eye Biomarkers of Cognitive Impairment by Measuring Vascular and Neurogenic Changes in the Retina , 2018, Front. Physiol..

[48]  James R. Burke,et al.  Retinal Microvascular and Neurodegenerative Changes in Alzheimer's Disease and Mild Cognitive Impairment Compared with Control Participants. , 2019, Ophthalmology. Retina.

[49]  K. Ghasemi Falavarjani,et al.  Image artefacts in swept-source optical coherence tomography angiography , 2016, British Journal of Ophthalmology.

[50]  Yongtian Wang,et al.  Automatic 2-D/3-D Vessel Enhancement in Multiple Modality Images Using a Weighted Symmetry Filter , 2018, IEEE Transactions on Medical Imaging.

[51]  P. Scheltens,et al.  Research criteria for the diagnosis of Alzheimer's disease: revising the NINCDS–ADRDA criteria , 2007, The Lancet Neurology.

[52]  David J. Wilson,et al.  Detailed Vascular Anatomy of the Human Retina by Projection-Resolved Optical Coherence Tomography Angiography , 2017, Scientific Reports.

[53]  Krzysztof Krawiec,et al.  Segmenting Retinal Blood Vessels With Deep Neural Networks , 2016, IEEE Transactions on Medical Imaging.

[54]  Zhiming Luo,et al.  Weighted Res-UNet for High-Quality Retina Vessel Segmentation , 2018, 2018 9th International Conference on Information Technology in Medicine and Education (ITME).

[55]  David Huang,et al.  Automated detection of shadow artifacts in optical coherence tomography angiography. , 2019, Biomedical optics express.

[56]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[57]  Bunyarit Uyyanonvara,et al.  Blood vessel segmentation methodologies in retinal images - A survey , 2012, Comput. Methods Programs Biomed..

[58]  Alfredo Ruggeri,et al.  A Fast and Efficient Technique for the Automatic Tracing of Corneal Nerves in Confocal Microscopy , 2016, Translational vision science & technology.

[59]  Xin Yang,et al.  A Three-Stage Deep Learning Model for Accurate Retinal Vessel Segmentation , 2019, IEEE Journal of Biomedical and Health Informatics.