This website requires JavaScript.

OCTA-500: A Retinal Dataset for Optical Coherence Tomography Angiography Study

Mingchao LiKun HuangQiuzhuo Xu ...+6 Qiang Chen
Dec 2020
摘要
Optical coherence tomography angiography (OCTA) is a novel imaging modalitythat has been widely utilized in ophthalmology and neuroscience studies toobserve retinal vessels and microvascular systems. However, publicly availableOCTA datasets remain scarce. In this paper, we introduce the largest and mostcomprehensive OCTA dataset dubbed OCTA-500, which contains OCTA imaging undertwo fields of view (FOVs) from 500 subjects. The dataset provides rich imagesand annotations including two modalities (OCT/OCTA volumes), six types ofprojections, four types of text labels (age / gender / eye / disease) and seventypes of segmentation labels (large vessel/capillary/artery/vein/2D FAZ/3DFAZ/retinal layers). Then, we propose a multi-object segmentation task calledCAVF, which integrates capillary segmentation, artery segmentation, veinsegmentation, and FAZ segmentation under a unified framework. In addition, weoptimize the 3D-to-2D image projection network (IPN) to IPN-V2 to serve as oneof the segmentation baselines. Experimental results demonstrate that IPN-V2achieves an ~10% mIoU improvement over IPN on CAVF task. Finally, we furtherstudy the impact of several dataset characteristics: the training set size, themodel input (OCT/OCTA, 3D volume/2D projection), the baseline networks, and thediseases. The dataset and code are publicly available at:https://ieee-dataport.org/open-access/octa-500.
展开全部
图表提取

暂无人提供速读十问回答

论文十问由沈向洋博士提出,鼓励大家带着这十个问题去阅读论文,用有用的信息构建认知模型。写出自己的十问回答,还有机会在当前页面展示哦。

Q1论文试图解决什么问题?
Q2这是否是一个新的问题?
Q3这篇文章要验证一个什么科学假设?
0
被引用
笔记
问答