验证视网膜眼底图像分析算法：问题与建议。
Validating retinal fundus image analysis algorithms: issues and a proposal.
机构信息
VAMPIRE project, School of Computing, University of Dundee, Dundee, United Kingdom.
出版信息
Invest Ophthalmol Vis Sci. 2013 May 1;54(5):3546-59. doi: 10.1167/iovs.12-10347.
Abstract
This paper concerns the validation of automatic retinal image analysis (ARIA) algorithms. For reasons of space and consistency, we concentrate on the validation of algorithms processing color fundus camera images, currently the largest section of the ARIA literature. We sketch the context (imaging instruments and target tasks) of ARIA validation, summarizing the main image analysis and validation techniques. We then present a list of recommendations focusing on the creation of large repositories of test data created by international consortia, easily accessible via moderated Web sites, including multicenter annotations by multiple experts, specific to clinical tasks, and capable of running submitted software automatically on the data stored, with clear and widely agreed-on performance criteria, to provide a fair comparison.
摘要
本文涉及自动视网膜图像分析 (ARIA) 算法的验证。由于篇幅和一致性的原因，我们专注于验证处理彩色眼底相机图像的算法，这是目前 ARIA 文献中最大的部分。我们概述了 ARIA 验证的背景（成像仪器和目标任务），总结了主要的图像分析和验证技术。然后，我们提出了一系列建议，重点是创建由国际联盟创建的大型测试数据存储库，这些存储库可通过经过审核的网站轻松访问，包括由多位专家针对特定临床任务进行的多中心注释，并且能够自动在存储的数据上运行提交的软件，同时具有明确且广泛达成共识的性能标准，以提供公平的比较。
相似文献
1
Validating retinal fundus image analysis algorithms: issues and a proposal.验证视网膜眼底图像分析算法：问题与建议。
Invest Ophthalmol Vis Sci. 2013 May 1;54(5):3546-59. doi: 10.1167/iovs.12-10347.
2
Assessment of image quality on color fundus retinal images using the automatic retinal image analysis.使用自动视网膜图像分析评估彩色眼底视网膜图像的质量。
Sci Rep. 2022 Jun 21;12(1):10455. doi: 10.1038/s41598-022-13919-2.
3
[Content-based automatic retinal image recognition and retrieval system].
Sheng Wu Yi Xue Gong Cheng Xue Za Zhi. 2013 Apr;30(2):403-8.
4
Retinal image enhancement based on color dominance of image.基于颜色优势的视网膜图像增强。
Sci Rep. 2023 May 3;13(1):7172. doi: 10.1038/s41598-023-34212-w.
5
Segmentation of the optic disk in color eye fundus images using an adaptive morphological approach.彩色眼底图像中视神经盘的自适应形态学分割。
Comput Biol Med. 2010 Feb;40(2):124-37. doi: 10.1016/j.compbiomed.2009.11.009. Epub 2009 Dec 31.
6
Regional Image Features Model for Automatic Classification between Normal and Glaucoma in Fundus and Scanning Laser Ophthalmoscopy (SLO) Images.用于眼底和扫描激光检眼镜（SLO）图像中正常与青光眼自动分类的区域图像特征模型
J Med Syst. 2016 Jun;40(6):132. doi: 10.1007/s10916-016-0482-9. Epub 2016 Apr 16.
7
Computer-assisted, interactive fundus image processing for macular drusen quantitation.用于黄斑玻璃膜疣定量分析的计算机辅助交互式眼底图像处理
Ophthalmology. 1999 Jun;106(6):1119-25. doi: 10.1016/S0161-6420(99)90257-9.
8
Automatic fundus image classification for computer-aided diagonsis.用于计算机辅助诊断的自动眼底图像分类
Annu Int Conf IEEE Eng Med Biol Soc. 2009;2009:1453-6. doi: 10.1109/IEMBS.2009.5332917.
9
DrishtiCare: a telescreening platform for diabetic retinopathy powered with fundus image analysis.DrishtiCare：一个基于眼底图像分析的糖尿病视网膜病变远程筛查平台。
J Diabetes Sci Technol. 2011 Jan 1;5(1):23-31. doi: 10.1177/193229681100500104.
10
Development and Validation of Deep Learning Models for Screening Multiple Abnormal Findings in Retinal Fundus Images.深度学习模型在视网膜眼底图像多种异常发现筛查中的开发与验证。
Ophthalmology. 2020 Jan;127(1):85-94. doi: 10.1016/j.ophtha.2019.05.029. Epub 2019 May 31.
引用本文的文献
1
SCONe: a community-acquired retinal image repository enabling ocular, cardiovascular and neurodegenerative disease prediction.SCONe：一个社区获得性视网膜图像库，可用于眼部、心血管和神经退行性疾病预测。
BMJ Health Care Inform. 2025 May 14;32(1):e101236. doi: 10.1136/bmjhci-2024-101236.
2
Latent space autoencoder generative adversarial model for retinal image synthesis and vessel segmentation.用于视网膜图像合成与血管分割的潜在空间自动编码器生成对抗模型。
BMC Med Imaging. 2025 May 5;25(1):149. doi: 10.1186/s12880-025-01694-1.
3
Automated analysis of vessel morphometry in retinal images from a Danish high street optician setting.丹麦高街配镜师环境下视网膜图像中血管形态计量的自动分析。
PLoS One. 2023 Aug 24;18(8):e0290278. doi: 10.1371/journal.pone.0290278. eCollection 2023.
4
Artificial intelligence in ophthalmology: The path to the real-world clinic.人工智能在眼科学中的应用：通往现实临床的道路。
Cell Rep Med. 2023 Jul 18;4(7):101095. doi: 10.1016/j.xcrm.2023.101095. Epub 2023 Jun 28.
5
Editorial: Functional assessments of the ocular circulation.社论：眼循环的功能评估
Front Med (Lausanne). 2023 Jun 8;10:1222022. doi: 10.3389/fmed.2023.1222022. eCollection 2023.
6
Artificial intelligence in retinal disease: clinical application, challenges, and future directions.人工智能在视网膜疾病中的应用：临床应用、挑战及未来方向。
Graefes Arch Clin Exp Ophthalmol. 2023 Nov;261(11):3283-3297. doi: 10.1007/s00417-023-06052-x. Epub 2023 May 9.
7
The impact of the image conversion factor and image centration on retinal vessel geometric characteristics.图像转换因子和图像对中对视网膜血管几何特征的影响。
Front Med (Lausanne). 2023 Mar 17;10:1112652. doi: 10.3389/fmed.2023.1112652. eCollection 2023.
8
Chákṣu: A glaucoma specific fundus image database.茶苦素：一种青光眼专用眼底图像数据库。
Sci Data. 2023 Feb 3;10(1):70. doi: 10.1038/s41597-023-01943-4.
9
Quality assessment of colour fundus and fluorescein angiography images using deep learning.利用深度学习对眼底彩照和荧光素血管造影图像进行质量评估。
Br J Ophthalmol. 2023 Dec 18;108(1):98-104. doi: 10.1136/bjo-2022-321963.
10
Multiple Preprocessing Hybrid Level Set Model for Optic Disc Segmentation in Fundus Images.多预处理混合水平集模型在眼底图像视盘分割中的应用。
Sensors (Basel). 2022 Sep 13;22(18):6899. doi: 10.3390/s22186899.
本文引用的文献
1
Estimating maximal measurable performance for automated decision systems from the characteristics of the reference standard. application to diabetic retinopathy screening.根据参考标准的特征估计自动化决策系统的最大可测量性能。应用于糖尿病视网膜病变筛查。
Annu Int Conf IEEE Eng Med Biol Soc. 2014;2014:154-7. doi: 10.1109/EMBC.2014.6943552.
2
Automated classification of severity of age-related macular degeneration from fundus photographs.基于眼底照片的年龄相关性黄斑变性严重程度的自动分类。
Invest Ophthalmol Vis Sci. 2013 Mar 11;54(3):1789-96. doi: 10.1167/iovs.12-10928.
3
Efficient optic cup detection from intra-image learning with retinal structure priors.基于视网膜结构先验的图像内学习实现高效视杯检测。
Med Image Comput Comput Assist Interv. 2012;15(Pt 1):58-65. doi: 10.1007/978-3-642-33415-3_8.
4
Does retinal vascular geometry vary with cardiac cycle?视网膜血管形态是否随心动周期变化？
Invest Ophthalmol Vis Sci. 2012 Aug 24;53(9):5799-805. doi: 10.1167/iovs.11-9326.
5
Depth discontinuity-based cup segmentation from multiview color retinal images.基于深度不连续的多视图彩色视网膜图像杯分割。
IEEE Trans Biomed Eng. 2012 Jun;59(6):1523-31. doi: 10.1109/TBME.2012.2187293. Epub 2012 Feb 10.
6
Retinal imaging and image analysis.视网膜成像与图像分析。
IEEE Rev Biomed Eng. 2010;3:169-208. doi: 10.1109/RBME.2010.2084567.
7
An automated retinal image quality grading algorithm.一种自动化的视网膜图像质量分级算法。
Annu Int Conf IEEE Eng Med Biol Soc. 2011;2011:5955-8. doi: 10.1109/IEMBS.2011.6091472.
8
Microaneurysm detection with radon transform-based classification on retina images.基于拉东变换分类的视网膜图像微动脉瘤检测
Annu Int Conf IEEE Eng Med Biol Soc. 2011;2011:5939-42. doi: 10.1109/IEMBS.2011.6091562.
9
Automatic glaucoma diagnosis from fundus image.
Annu Int Conf IEEE Eng Med Biol Soc. 2011;2011:3383-6. doi: 10.1109/IEMBS.2011.6090916.
10
Sliding window and regression based cup detection in digital fundus images for glaucoma diagnosis.基于滑动窗口和回归的数字眼底图像杯盘检测用于青光眼诊断。
Med Image Comput Comput Assist Interv. 2011;14(Pt 3):1-8. doi: 10.1007/978-3-642-23626-6_1.
Zotero 插件