Optical coherence tomography angiography (OCTA) offers a noninvasive label-free solution for imaging retinal vasculatures at the capillary level resolution. In principle, improved resolution implies a better chance to reveal subtle microvascular distortions associated with eye diseases that are asymptomatic in early stages. However, massive screening requires experienced clinicians to manually examine retinal images, which may result in human error and hinder objective screening. Recently, quantitative OCTA features have been developed to standardize and document retinal vascular changes. The feasibility of using quantitative OCTA features for machine learning classification of different retinopathies has been demonstrated. Deep learning-based applications have also been explored for automatic OCTA image analysis and disease classification. In this article, we summarize recent developments of quantitative OCTA features, machine learning image analysis, and classification.
TingDSLiuYBurlinaPXuXBresslerNMWongTY.AI for medical imaging goes deep. Nat Med2018;
24:539–40
2.
KharratAHalimaMBAyedMB. MRI brain tumor classification using support vector machines and meta-heuristic method. In: 2015 15th international conference on intelligent systems design and applications (ISDA), Marrakech, Morocco, 2015, pp. 446–51. Piscataway: IEEE.
3.
MohsenHEl-DahshanE-SAEl-HorbatyE-SMSalemA-BM.Classification using deep learning neural networks for brain tumors. Future Comput Inform J2018;
3:68–71
4.
ChatoLLatifiS. Machine learning and deep learning techniques to predict overall survival of brain tumor patients using MRI images. In: 2017 IEEE 17th international conference on bioinformatics and bioengineering (BIBE), Washington, DC, USA, 2017, pp.9–14. Piscataway: IEEE.
5.
BeckerASMuellerMStoffelEMarconMGhafoorSBossA.Classification of breast cancer in ultrasound imaging using a generic deep learning analysis software: a pilot study. Br J Radiol2018;
91:20170576
6.
TanakaHChiuS-WWatanabeTKaokuSYamaguchiT.Computer-aided diagnosis system for breast ultrasound images using deep learning. Phys Med Biol2019;
64:235013
7.
HanSKangH-KJeongJ-YParkM-HKimWBangW-CSeongY-K.A deep learning framework for supporting the classification of breast lesions in ultrasound images. Phys Med Biol2017;
62:7714
8.
LakhaniPSundaramB.Deep learning at chest radiography: automated classification of pulmonary tuberculosis by using convolutional neural networks. Radiology2017;
284:574–82
9.
BaltruschatIMNickischHGrassMKnoppTSaalbachA.Comparison of deep learning approaches for multi-label chest X-ray classification. Sci Rep2019;
9:1–10
10.
GuptaAGuptaSKataryaR.InstaCovNet-19: a deep learning classification model for the detection of COVID-19 patients using chest X-ray. Appl Soft Comput2021;
99:106859
11.
HigakiTNakamuraYZhouJYuZNemotoTTatsugamiFAwaiK.Deep learning reconstruction at CT: phantom study of the image characteristics. Acad Radiol2020;
27:82–7
12.
BhatiaSSinhaYGoelL. Lung cancer detection: a deep learning approach. In: Bansal JC, Das KN, Nagar A, Deep K, Ojha AK (eds) Soft computing for problem solving. Berlin: Springer, 2019, pp. 699–705
13.
RiquelmeDAkhloufiMA.Deep learning for lung cancer nodules detection and classification in CT scans. AI2020;
1:28–67
14.
ChewEYSchachatAP.Should we add screening of age-related macular degeneration to current screening programs for diabetic retinopathy?Ophthalmology2015;
122:2155–6
15.
AlamMLeDLimJIYaoX.Vascular complexity analysis in optical coherence tomography angiography of diabetic retinopathy. Retina2021;
41:538–45
16.
TingDSWCheungCY-LLimGTanGSWQuangNDGanAHamzahHGarcia-FrancoRSan YeoIYLeeSY.Development and validation of a deep learning system for diabetic retinopathy and related eye diseases using retinal images from multiethnic populations with diabetes. JAMA2017;
318:2211–23
17.
BurlinaPMJoshiNPekalaMPachecoKDFreundDEBresslerNM.Automated grading of age-related macular degeneration from color fundus images using deep convolutional neural networks. JAMA Ophthalmol2017;
135:1170–6
18.
LeeCSBaughmanDMLeeAY.Deep learning is effective for classifying normal versus age-related macular degeneration OCT images. Ophthalmol Retina2017;
1:322–7
19.
AlamMThapaDLimJICaoDYaoX.Computer-aided classification of sickle cell retinopathy using quantitative features in optical coherence tomography angiography. Biomed Opt Express2017;
8:4206–16
20.
AlamMLeDLimJIChanRVYaoX.Supervised machine learning based multi-task artificial intelligence classification of retinopathies. J Clin Med2019;
8:872
21.
NagasatoDTabuchiHOhsugiHMasumotoHEnnoHIshitobiNSonobeTKameokaMNikiMHayashiK.Deep neural network-based method for detecting Central retinal vein occlusion using ultrawide-field fundus ophthalmoscopy. J Ophthalmol2018;
2018:https://doi.org/10.1155/2018/1875431
MookiahMRKAcharyaURChuaCKLimCMNgELaudeA.Computer-aided diagnosis of diabetic retinopathy: a review. Comput Biol Med2013;
43:2136–55
24.
ShahSAALaudeAFayeITangTB.Automated microaneurysm detection in diabetic retinopathy using curvelet transform. J Biomed Opt2016;
21:101404
25.
WinderRJMorrowPJMcRitchieINBailieJHartPM.Algorithms for digital image processing in diabetic retinopathy. Comput Med Imaging Graph2009;
33:608–22
26.
BhattacharyyaSKonarDPlatosJKarCSharmaK.Hybrid machine intelligence for medical image analysis.
Berlin:
Springer, 2020
27.
EladawiNElmogyMKhalifaFGhazalMGhaziNAboelfetouhARiadASandhuHSchaalSEl‐BazA.Early diabetic retinopathy diagnosis based on local retinal blood vessel analysis in optical coherence tomography angiography (OCTA) images. Med Phys2018;
45:4582–99
28.
SandhuHSEladawiNElmogyMKeyntonRHelmyOSchaalSEl-BazA.Automated diabetic retinopathy detection using optical coherence tomography angiography: a pilot study. Br J Ophthalmol2018;
102:1564–9
29.
ThompsonACJammalAAMedeirosFAA.Review of deep learning for screening, diagnosis, and detection of glaucoma progression. Transl Vis Sci Technol2020;
9:42
30.
AlamMZhangYLimJIChanRVYangMYaoX.Quantitative optical coherence tomography angiography features for objective classification and staging of diabetic retinopathy. Retina2020;
40:322–32
31.
ShiiharaHTerasakiHSonodaSKakiuchiNShinoharaYTomitaMSakamotoT.Objective evaluation of size and shape of superficial foveal avascular zone in normal subjects by optical coherence tomography angiography. Sci Rep2018;
8:1–9
32.
TakaseNNozakiMKatoAOzekiHYoshidaMOguraY.Enlargement of foveal avascular zone in diabetic eyes evaluated by en face optical coherence tomography angiography. Retina2015;
35:2377–83
33.
KashaniAHLeeSYMoshfeghiADurbinMKPuliafitoCA.Optical coherence tomography angiography of retinal venous occlusion. Retina2015;
35:2323–31
34.
WatsonAB.Perimetric complexity of binary digital images. Math J2012;
14:1–40
35.
Al-KadiOSWatsonD.Texture analysis of aggressive and nonaggressive lung tumor CE CT images. IEEE Trans Biomed Eng2008;
55:1822–30
36.
AslamTMHoyleDCPuriVBentoG.Differentiation of diabetic status using statistical and machine learning techniques on optical coherence tomography angiography images. Transl Vis Sci Technol2020;
9:2–2
37.
SandhuHSElmogyMSharafeldeenATElsharkawyMEl-AdawyNEltanbolyAShalabyAKeyntonREl-BazA.Automated diagnosis of diabetic retinopathy using clinical biomarkers, optical coherence tomography, and optical coherence tomography angiography. Am J Ophthalmol2020;
216:201–06
38.
LeeCSTyringAJWuYXiaoSRokemASDeruyterNPZhangQTufailAWangRKLeeAY.Generating retinal flow maps from structural optical coherence tomography with artificial intelligence. Sci Rep2019;
9:1–11
39.
AlamMThapaDLimJICaoDYaoX.Quantitative characteristics of sickle cell retinopathy in optical coherence tomography angiography. Biomed Opt Express2017;
8:1741–53
40.
JiangZHuangZQiuBMengXYouYLiuXLiuGZhouCYangKMaierA.Comparative study of deep learning models for optical coherence tomography angiography. Biomed Opt Express2020;
11:1580–97
41.
LiPLO'NeilCSaberiSSinderKWangKTanBHosseinaeeZBizhevatKLakshminarayananV. Deep learning algorithm for generating optical coherence tomography angiography (OCTA) maps of the retinal vasculature. Appl Mach Learn 2020;11511:1151109.
42.
KadomotoSUjiAMuraokaYAkagiTTsujikawaA.Enhanced visualization of retinal microvasculature in optical coherence tomography angiography imaging via deep learning. J Clin Med2020;
9:1322
43.
GaoMGuoYHormelTTSunJHwangTSJiaY.Reconstruction of high-resolution 6× 6-mm OCT angiograms using deep learning. Biomed Opt Express2020;
11:3585–600
44.
GaoDCelikNWuXWilliamsBMStylianidesAZhengY. A novel deep learning based OCTA de-striping method. In: Annual conference on medical image understanding and analysis. Berlin: Springer, 2019, pp. 189–97
45.
WangJHormelTTGaoLZangPGuoYWangXBaileySTJiaY.Automated diagnosis and segmentation of choroidal neovascularization in OCT angiography using deep learning. Biomed Opt Express2020;
11:927–44
46.
MaYHaoHXieJFuHZhangJYangJWangZLiuJZhengYZhaoY.ROSE: a retinal OCT-angiography vessel segmentation dataset and new model. IEEE Trans Med Imaging 2021;40:928–39
47.
MirshahiRAnvariPRiazi-EsfahaniHSardariniaMNaseripourMFalavarjaniKG.Foveal avascular zone segmentation in optical coherence tomography angiography images using a deep learning approach. Sci Rep2021;
11:1–8
48.
GuoYHormelTTXiongHWangJHwangTSJiaY.Automated segmentation of retinal fluid volumes from structural and angiographic optical coherence tomography using deep learning. Transl Vis Sci Technol2020;
9:54–54
49.
JoukalM.Anatomy of the human visual pathway. Homonymous visual field defects.
Berlin:
Springer, 2017, pp. 1–16
50.
SimonyanKZissermanA. Very deep convolutional networks for large-scale image recognition. arXiv preprint 2014;1409:1556.
51.
DengJDongWSocherRLiL-JLiKFei-FeiL. Imagenet: a large-scale hierarchical image database. In: 2009 IEEE conference on computer vision and pattern recognition, June 20, 2009, pp. 248–55. Piscataway: IEEE.
52.
HeislerMKarstSLoJMammoZYuTWarnerSMaberleyDBegMFNavajasEVSarunicMV.Ensemble deep learning for diabetic retinopathy detection using optical coherence tomography angiography. Transl Vis Sci Technol2020;
9:20
53.
FonsecaRADantasMA.Retinal venous beading associated with recurrent branch vein occlusion. Can J Ophthalmol2002;
37:182–83
54.
PedersenLJeppesenPKnudsenSTPoulsenPLBekT.Improvement of mild retinopathy in type 2 diabetic patients correlates with narrowing of retinal arterioles. A prospective observational study. Graefes ArchClin Exp Ophthalmol2014;
252:1561–67
55.
CheungNBluemkeDAKleinRSharrettARIslamFACotchMFKleinBECriquiMHWongTY.Retinal arteriolar narrowing and left ventricular remodeling: the multi-ethnic study of atherosclerosis. J Am Coll Cardiol2007;
50:48–55
AlamMLimJIToslakDYaoX.Differential artery–vein analysis improves the performance of OCTA staging of sickle cell retinopathy. Trans Vis Sci Tech2019;
8:3–3
AlamMToslakDLimJIYaoX.Color fundus image guided artery-vein differentiation in optical coherence tomography angiography. Invest Ophthalmol Vis Sci2018;
59:4953–62
60.
LeDAlamMYaoCKLimJIHsiehY-TChanRVToslakDYaoX.Transfer learning for automated OCTA detection of diabetic retinopathy. Transl Vis Sci Technol2020;
9:35–35
61.
AlamMLeDSonTLimJIYaoX.AV-Net: deep learning for fully automated artery-vein classification in optical coherence tomography angiography. Biomed Opt Express2020;
11:5249–57
62.
NesperPLRobertsPKOnishiACChaiHLiuLJampolLMFawziAA.Quantifying microvascular abnormalities with increasing severity of diabetic retinopathy using optical coherence tomography angiography. Invest Ophthalmol Vis Sci2017;
58:BIO307–BIO15
63.
GuoYCaminoAWangJHuangDHwangTSJiaY.MEDnet, a neural network for automated detection of avascular area in OCT angiography. Biomed Opt Express2018;
9:5147–58
64.
NagasatoDTabuchiHMasumotoHEnnoHIshitobiNKameokaMNikiMMitamuraY.Automated detection of a nonperfusion area caused by retinal vein occlusion in optical coherence tomography angiography images using deep learning. PLoS One2019;
14:e0223965
65.
KermanyDSGoldbaumMCaiWValentimCCLiangHBaxterSLMcKeownAYangGWuXYanF.Identifying medical diagnoses and treatable diseases by image-based deep learning. Cell2018;
172:1122–31.e9
