ROSE: Retinal OCT-Angiography segmentation dataset

<p>Retinal OCT-Angiography Vessel Segmentation Dataset (ROSE) is now open source. It includes two subsets:&nbsp;<strong>ROSE-1</strong>&nbsp;and&nbsp;<strong>ROSE-2</strong>.&nbsp;</p> <ul> <li><strong>ROSE-1</strong>&nbsp;subset consists of a total of&nbsp;<strong>117&nbsp;</strong>OCTA images from 39 subjects (26 with disease &nbsp;and the rest are healthy control) and was split into 90 images for training and 27 images for testing. Each subject has&nbsp;<em>en face</em>&nbsp;angiograms of superficial (SVC), deep (DVC), and the inner retinal vascular plexus that includes both SVC and DVC (SVC+DVC)&nbsp;respectively. All the OCTA scans were captured by the&nbsp;<u>RTVue XR Avanti SD-OCT system (Optovue, USA) equipped with AngioVue software</u>, with image resolution of 304 &times; 304 pixels. The scan area was 3&times;3 mm²&nbsp;area centred at the fovea. The manual annotations of these vascular network were graded &nbsp;by image experts and clinicians, and the consensus of them was then used as the groundtruth.&nbsp;Two different types of vessel annotations are available:&nbsp;<strong>centerline-level</strong>&nbsp;and&nbsp;<strong>pixel-level</strong>&nbsp;annotation.&nbsp;</li> <li><strong>ROSE-2</strong>&nbsp;subset contains a total of&nbsp;<strong>112</strong>&nbsp;OCT-A images of 112 eyes acquired by&nbsp;<u>Heidelberg OCT2 system with Spectralis software (Heidelberg Engineering, Heidelberg, Germany)&nbsp;</u>and was split into 90 images for training and 22 images for testing. All the images in this dataset are the en face angiograms of SVC within a 3&times;3 mm² area centred at the fovea, and each image was resized into a grayscale image with 840 &times; 840 pixels. All the visible retinal vessels were manually traced using an in-house program written in Matlab (Mathworks R2018, Natwick) by an experienced clinical ophthalmologist.&nbsp;</li> </ul> <p>code: <a href="https://github.com/iMED-Lab/OCTA-Net-OCTA-Vessel-Segmentation-Network">iMED-Lab/OCTA-Net-OCTA-Vessel-Segmentation-Network (github.com)</a></p> <p><strong>ROSE-O</strong>&nbsp;is a retinal structure detection dataset of OCTA images, with precise manual annotations of RV, RVJ and the FAZ. It contains 117 images which were captured using the&nbsp;<u>Optovue Avanti RTVue XR with AngioVue software (Optovue, Fremont, USA)</u>: the images have a resolution of 304&times;304 pixels. The SVC, DVC and IVC angiograms of each participant were obtained by the device.&nbsp;</p> <p>code: <a href="https://github.com/iMED-Lab/VAFF-Net">https://github.com/iMED-Lab/VAFF-Net</a></p>