用于在光学相干断层扫描血管造影中区分无灌注区与信号衰减伪影的深度学习算法的开发与验证
Development and validation of a deep learning algorithm for distinguishing the nonperfusion area from signal reduction artifacts on OCT angiography.
作者信息
Guo Yukun, Hormel Tristan T, Xiong Honglian, Wang Bingjie, Camino Acner, Wang Jie, Huang David, Hwang Thomas S, Jia Yali
机构信息
Casey Eye Institute, Oregon Health & Science University, Portland, OR 97239, USA.
These authors contributed equally.
出版信息
Biomed Opt Express. 2019 Jun 12;10(7):3257-3268. doi: 10.1364/BOE.10.003257. eCollection 2019 Jul 1.
Abstract
The capillary nonperfusion area (NPA) is a key quantifiable biomarker in the evaluation of diabetic retinopathy (DR) using optical coherence tomography angiography (OCTA). However, signal reduction artifacts caused by vitreous floaters, pupil vignetting, or defocus present significant obstacles to accurate quantification. We have developed a convolutional neural network, MEDnet-V2, to distinguish NPA from signal reduction artifacts in 6×6 mm OCTA. The network achieves strong specificity and sensitivity for NPA detection across a wide range of DR severity and scan quality.
摘要
毛细血管无灌注区（NPA）是使用光学相干断层扫描血管造影（OCTA）评估糖尿病视网膜病变（DR）时的一个关键可量化生物标志物。然而，由玻璃体混浊、瞳孔渐晕或散焦引起的信号降低伪影对准确量化构成了重大障碍。我们开发了一种卷积神经网络MEDnet-V2，用于在6×6 mm的OCTA中区分NPA和信号降低伪影。该网络在广泛的DR严重程度和扫描质量范围内，对NPA检测具有很强的特异性和敏感性。
相似文献
1
Development and validation of a deep learning algorithm for distinguishing the nonperfusion area from signal reduction artifacts on OCT angiography.用于在光学相干断层扫描血管造影中区分无灌注区与信号衰减伪影的深度学习算法的开发与验证
Biomed Opt Express. 2019 Jun 12;10(7):3257-3268. doi: 10.1364/BOE.10.003257. eCollection 2019 Jul 1.
2
Quantification of Nonperfusion Area in Montaged Widefield OCT Angiography Using Deep Learning in Diabetic Retinopathy.在糖尿病视网膜病变中利用深度学习对拼接式超广角光学相干断层扫描血管造影中的无灌注区进行定量分析。
Ophthalmol Sci. 2021 May 12;1(2):100027. doi: 10.1016/j.xops.2021.100027. eCollection 2021 Jun.
3
Robust non-perfusion area detection in three retinal plexuses using convolutional neural network in OCT angiography.在光学相干断层扫描血管造影中使用卷积神经网络对三个视网膜神经丛进行稳健的无灌注区检测。
Biomed Opt Express. 2019 Dec 18;11(1):330-345. doi: 10.1364/BOE.11.000330. eCollection 2020 Jan 1.
4
Automated detection of a nonperfusion area caused by retinal vein occlusion in optical coherence tomography angiography images using deep learning.利用深度学习技术在光学相干断层扫描血管造影图像中自动检测视网膜静脉阻塞引起的无灌注区。
PLoS One. 2019 Nov 7;14(11):e0223965. doi: 10.1371/journal.pone.0223965. eCollection 2019.
5
Automated Quantification of Capillary Nonperfusion Using Optical Coherence Tomography Angiography in Diabetic Retinopathy.利用光学相干断层扫描血管造影术自动定量分析糖尿病视网膜病变中的毛细血管无灌注情况。
JAMA Ophthalmol. 2016 Apr;134(4):367-73. doi: 10.1001/jamaophthalmol.2015.5658.
6
QUANTIFICATION OF RETINAL CAPILLARY NONPERFUSION IN DIABETICS USING WIDE-FIELD OPTICAL COHERENCE TOMOGRAPHY ANGIOGRAPHY.应用广角光学相干断层扫描血管造影术定量检测糖尿病患者视网膜毛细血管无灌注。
Retina. 2020 Mar;40(3):412-420. doi: 10.1097/IAE.0000000000002403.
7
Automated detection of shadow artifacts in optical coherence tomography angiography.光学相干断层扫描血管造影中阴影伪影的自动检测
Biomed Opt Express. 2019 Feb 28;10(3):1514-1531. doi: 10.1364/BOE.10.001514. eCollection 2019 Mar 1.
8
Ultra-Widefield Protocol Enhances Automated Classification of Diabetic Retinopathy Severity with OCT Angiography.超广角协议结合 OCT 血管造影增强了糖尿病视网膜病变严重程度的自动分类。
Ophthalmol Retina. 2020 Apr;4(4):415-424. doi: 10.1016/j.oret.2019.10.018. Epub 2019 Nov 9.
9
MEDnet, a neural network for automated detection of avascular area in OCT angiography.MEDnet，一种用于在光学相干断层扫描血管造影中自动检测无血管区域的神经网络。
Biomed Opt Express. 2018 Oct 2;9(11):5147-5158. doi: 10.1364/BOE.9.005147. eCollection 2018 Nov 1.
10
Physics-Based Optical Coherence Tomography Angiography (OCTA) Image Correction for Shadow Compensation.基于物理的光学相干断层扫描血管造影（OCTA）图像校正以进行阴影补偿
IEEE Trans Biomed Eng. 2025 Mar;72(3):891-898. doi: 10.1109/TBME.2024.3478384. Epub 2025 Feb 20.
引用本文的文献
1
Quantifying Choriocapillaris Flow Deficits in Diabetic Retinopathy Using Projection-Resolved OCT Angiography.使用投影分辨光学相干断层扫描血管造影术量化糖尿病视网膜病变中的脉络膜毛细血管血流缺损
Invest Ophthalmol Vis Sci. 2025 Sep 2;66(12):13. doi: 10.1167/iovs.66.12.13.
2
Wide-field OCTA quantified peripheral nonperfusion areas predict the risk of subclinical neovascularization.广域光学相干断层扫描血管造影定量分析的周边无灌注区可预测亚临床新生血管形成的风险。
Eye (Lond). 2025 Jul 2. doi: 10.1038/s41433-025-03891-2.
3
Widefield OCT angiography.广角光学相干断层扫描血管造影术
Prog Retin Eye Res. 2025 Jul;107:101378. doi: 10.1016/j.preteyeres.2025.101378. Epub 2025 Jun 13.
4
Wide-field OCTA Quantified Peripheral Nonperfusion Areas Predict the Risk of Subclinical Neovascularization.广角光学相干断层扫描血管造影定量分析外周无灌注区可预测亚临床新生血管形成的风险。
Res Sq. 2025 May 5:rs.3.rs-6182918. doi: 10.21203/rs.3.rs-6182918/v1.
5
Recent advances and applications of optical coherence tomography angiography in diabetic retinopathy.光学相干断层扫描血管造影在糖尿病视网膜病变中的最新进展与应用
Front Endocrinol (Lausanne). 2025 Apr 16;16:1438739. doi: 10.3389/fendo.2025.1438739. eCollection 2025.
6
Nonperfused Retinal Capillaries-A New Method Developed on OCT and OCTA.无灌注视网膜毛细血管——一种基于光学相干断层扫描（OCT）和光学相干断层扫描血管造影（OCTA）开发的新方法。
Invest Ophthalmol Vis Sci. 2025 Apr 1;66(4):22. doi: 10.1167/iovs.66.4.22.
7
Artificial Intelligence Versus Rules-Based Approach for Segmenting NonPerfusion Area in a DRCR Retina Network Optical Coherence Tomography Angiography Dataset.在DRCR视网膜网络光学相干断层扫描血管造影数据集中，用于分割无灌注区的人工智能与基于规则的方法对比
Invest Ophthalmol Vis Sci. 2025 Mar 3;66(3):22. doi: 10.1167/iovs.66.3.22.
8
Quantitative Volumetric Analysis of Retinal Ischemia with an Oxygen Diffusion Model and OCT Angiography.基于氧扩散模型和光学相干断层扫描血管造影术的视网膜缺血定量容积分析
Ophthalmol Sci. 2024 Jul 19;4(6):100579. doi: 10.1016/j.xops.2024.100579. eCollection 2024 Nov-Dec.
9
Multi-Plexus Nonperfusion Area Segmentation in Widefield OCT Angiography Using a Deep Convolutional Neural Network.使用深度卷积神经网络对广角 OCT 血管造影中的多丛无灌注区进行分割。
Transl Vis Sci Technol. 2024 Jul 1;13(7):15. doi: 10.1167/tvst.13.7.15.
10
Evaluation of optical coherence tomography angiography metrics in children and adolescents with type 1 diabetes: 4-year longitudinal study.1 型糖尿病儿童和青少年光学相干断层扫描血管造影指标的评估：4 年纵向研究。
Acta Diabetol. 2024 Oct;61(10):1211-1223. doi: 10.1007/s00592-024-02291-4. Epub 2024 May 3.
本文引用的文献
1
Generating retinal flow maps from structural optical coherence tomography with artificial intelligence.利用人工智能从结构光相干断层扫描生成视网膜血流图。
Sci Rep. 2019 Apr 5;9(1):5694. doi: 10.1038/s41598-019-42042-y.
2
Automated detection of shadow artifacts in optical coherence tomography angiography.光学相干断层扫描血管造影中阴影伪影的自动检测
Biomed Opt Express. 2019 Feb 28;10(3):1514-1531. doi: 10.1364/BOE.10.001514. eCollection 2019 Mar 1.
3
Deep learning is effective for the classification of OCT images of normal versus Age-related Macular Degeneration.深度学习对于正常与年龄相关性黄斑变性的光学相干断层扫描（OCT）图像分类很有效。
Ophthalmol Retina. 2017 Jul-Aug;1(4):322-327. doi: 10.1016/j.oret.2016.12.009. Epub 2017 Feb 13.
4
Automated segmentation of retinal layer boundaries and capillary plexuses in wide-field optical coherence tomographic angiography.超广角光学相干断层扫描血管造影中视网膜层边界和毛细血管丛的自动分割
Biomed Opt Express. 2018 Aug 24;9(9):4429-4442. doi: 10.1364/BOE.9.004429. eCollection 2018 Sep 1.
5
QUANTIFICATION OF RETINAL CAPILLARY NONPERFUSION IN DIABETICS USING WIDE-FIELD OPTICAL COHERENCE TOMOGRAPHY ANGIOGRAPHY.应用广角光学相干断层扫描血管造影术定量检测糖尿病患者视网膜毛细血管无灌注。
Retina. 2020 Mar;40(3):412-420. doi: 10.1097/IAE.0000000000002403.
6
MEDnet, a neural network for automated detection of avascular area in OCT angiography.MEDnet，一种用于在光学相干断层扫描血管造影中自动检测无血管区域的神经网络。
Biomed Opt Express. 2018 Oct 2;9(11):5147-5158. doi: 10.1364/BOE.9.005147. eCollection 2018 Nov 1.
7
Artificial intelligence and deep learning in ophthalmology.人工智能和深度学习在眼科学中的应用。
Br J Ophthalmol. 2019 Feb;103(2):167-175. doi: 10.1136/bjophthalmol-2018-313173. Epub 2018 Oct 25.
8
DRUNET: a dilated-residual U-Net deep learning network to segment optic nerve head tissues in optical coherence tomography images.DRUNET：一种用于在光学相干断层扫描图像中分割视神经乳头组织的扩张残差U型网络深度学习网络。
Biomed Opt Express. 2018 Jun 25;9(7):3244-3265. doi: 10.1364/BOE.9.003244. eCollection 2018 Jul 1.
9
Effect of patch size and network architecture on a convolutional neural network approach for automatic segmentation of OCT retinal layers.补丁大小和网络架构对用于光学相干断层扫描视网膜层自动分割的卷积神经网络方法的影响。
Biomed Opt Express. 2018 Jun 11;9(7):3049-3066. doi: 10.1364/BOE.9.003049. eCollection 2018 Jul 1.
10
Automated Quantification of Nonperfusion Areas in 3 Vascular Plexuses With Optical Coherence Tomography Angiography in Eyes of Patients With Diabetes.糖尿病患者眼中的光学相干断层血管造影 3 个血管丛无灌注区的自动量化。
JAMA Ophthalmol. 2018 Aug 1;136(8):929-936. doi: 10.1001/jamaophthalmol.2018.2257.
Zotero 插件