Applications of Artificial Intelligence in Optical Coherence Tomography Angiography Imaging

Abstract

Optical coherence tomography angiography (OCTA) and artificial intelligence (AI) are two emerging fields that complement each other. OCTA enables the noninvasive, in vivo, 3D visualization of retinal blood flow with a micrometer resolution, which has been impossible with other imaging modalities. As it does not need dye-based injections, it is also a safer procedure for patients. AI has excited great interest in many fields of daily life, by enabling automatic processing of huge amounts of data with a performance that greatly surpasses previous algorithms. It has been used in many breakthrough studies in recent years, such as the finding that AlphaGo can beat humans in the strategic board game of Go. This paper will give a short introduction into both fields and will then explore the manifold applications of AI in OCTA imaging that have been presented in the recent years. These range from signal generation over signal enhancement to interpretation tasks like segmentation and classification. In all these areas, AI-based algorithms have achieved state-of-the-art performance that has the potential to improve standard care in ophthalmology when integrated into the daily clinical routine.

Zusammenfassung

Die optische Kohärenztomografie-Angiografie (OCTA) und die künstliche Intelligenz (KI) sind 2 aufstrebende Bereiche, die sich gegenseitig ergänzen. Die OCTA ermöglicht die nicht invasive In-vivo-3-D-Visualisierung des Blutflusses in der Netzhaut mit einer Auflösung im Mikrometerbereich, was mit anderen Bildgebungsmodalitäten bisher nicht möglich war. Da keine Farbstoffinjektionen erforderlich sind, ist das Verfahren auch für die Patienten sicherer. Die KI hat in vielen Bereichen des täglichen Lebens großes Interesse geweckt, da sie die automatische Verarbeitung großer Datenmengen ermöglicht und die Leistung bisheriger Algorithmen weit übertrifft. Sie wurde in den letzten Jahren in vielen bahnbrechenden Arbeiten eingesetzt, wie z. B. AlphaGo, das im strategische Brettspiel Go Menschen übertrifft. Dieser Beitrag wird eine kurze Einführung in diese beiden Themen geben und dann die vielfältigen Anwendungen von KI in der OCTA-Bildgebung beleuchten, die in den letzten Jahren vorgestellt wurden. Diese reichen von der Signalgenerierung über die Signalverbesserung bis hin zu Interpretationsaufgaben wie Segmentierung und Klassifikation. In all diesen Bereichen haben KI-basierte Algorithmen Spitzenleistungen erzielt, die das Potenzial haben, die Standardversorgung in der Augenheilkunde zu verbessern, wenn sie in die tägliche klinische Routine integriert werden.

Key words

OCTA - artificial intelligence - machine learning - deep learning

Schlüsselwörter

OCTA - künstliche Intelligenz - maschinelles Lernen - tiefes Lernen

Publikationsverlauf

Eingereicht: 01. Juni 2022

Angenommen: 25. September 2022

Artikel online veröffentlicht:
09. Dezember 2022

• References

• 1 Knowles E. The Oxford Dictionary of Phrase and Fable. Oxford: Oxford University Press; 2006
• 2 Goodfellow I, Bengio Y, Courville A. Deep Learning. Cambridge, MA, USA: MIT Press; 2016
• 3 Buchanan BG, Shortliffe EH. Rule-based Expert Systems: The MYCIN Experiments of the Stanford Heuristic Programming Project. Reading, MA: Addison-Wesley; 1984
• 4 Russell SJ. Artificial Intelligence: A modern Approach. Pearson Education, Inc.; 2010
• 5 Bengio Y, Courville A, Vincent P. Representation learning: a review and new perspectives. IEEE Trans Pattern Anal Mach Intell 2013; 35: 1798-1828
• 6 Litjens G, Kooi T, Bejnordi BE. et al. A survey on deep learning in medical image analysis. Med Image Anal 2017; 42: 60-88
• 7 Hubel DH, Wiesel TN. Receptive fields of single neurones in the catʼs striate cortex. J Physiol 1959; 148: 574-591
• 8 He K, Zhang X, Ren S. et al. Deep Residual Learning for Image Recognition. In: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas, NV, USA: IEEE; 2016: 770-778
• 9 Ronneberger O, Fischer P, Brox T. U-Net: Convolutional Networks for Biomedical Image Segmentation. In: Navab N, Hornegger J, Wells WM, Frangi AF, eds. International Conference on Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015. Cham: Springer; 2015: 234-241
• 10 Goodfellow I, Pouget-Abadie J, Mirza M. et al. Generative Adversarial Networks. Communications of the ACM 2020; 63: 139-144
• 11 Campbell JP, Zhang M, Hwang TS. et al. Detailed Vascular Anatomy of the Human Retina by Projection-Resolved Optical Coherence Tomography Angiography. Sci Rep 2017; 7: 42201
• 12 Bailey ST, Thaware O, Wang J. et al. Detection of nonexudative choroidal neovascularization and progression to exudative choroidal neovascularization using OCT angiography. Ophthalmol Retina 2019; 3: 629-636
• 13 de Oliveira Dias JR, Zhang Q, Garcia JM. et al. Natural history of subclinical neovascularization in nonexudative age-related macular degeneration using swept-source OCT angiography. Ophthalmology 2018; 125: 255-266
• 14 Hwang TS, Gao SS, Liu L. et al. Automated Quantification of Capillary Nonperfusion Using Optical Coherence Tomography Angiography in Diabetic Retinopathy. JAMA Ophthalmol 2016; 134: 367-373
• 15 Ishibazawa A, Nagaoka T, Takahashi A. et al. Optical Coherence Tomography Angiography in Diabetic Retinopathy: A Prospective Pilot Study. Am J Ophthalmol 2015; 160: 35-44
• 16 Hormel TT, Jia YL, Jian YF. et al. Plexus-specific retinal vascular anatomy and pathologies as seen by projection-resolved optical coherence tomographic angiography. Prog Retin Eye Res 2021; 80: 100878
• 17 Patel RC, Wang J, Hwang TS. et al. Plexus-specific detection of retinal vascular pathologic conditions with projection-resolved OCT angiography. Ophthalmol Retina 2018; 2: 816-826
• 18 Schottenhamml J, Wurfl T, Mardin S. et al. Glaucoma classification in 3 × 3 mm en face macular scans using deep learning in a different plexus. Biomed Opt Express 2021; 12: 7434-7444
• 19 Heiferman M, Fawzi AA. Progression of subclinical choroidal neovascularization in age-related macular degeneration. PLoS One 2019; 14: e0217805
• 20 Rosen RB, Romo JSA, Krawitz BD. et al. Earliest Evidence of Preclinical Diabetic Retinopathy Revealed Using Optical Coherence Tomography Angiography Perfused Capillary Density. Am J Ophthalmol 2019; 203: 103-115
• 21 [Anonymous] Corrigendum. Erratum for: OCT Angiography Metrics Predict Progression of Diabetic Retinopathy and Development of Diabetic Macular Edema: A Prospective Study. Ophthalmology 2020; 127: 1777
• 22 Yanagi Y, Mohla A, Lee WK. et al. Prevalence and Risk Factors for Nonexudative Neovascularization in Fellow Eyes of Patients With Unilateral Age-Related Macular Degeneration and Polypoidal Choroidal Vasculopathy. Invest Ophthalmol Vis Sci 2017; 58: 3488-3495
• 23 de Carlo TE, Romano A, Waheed NK. et al. A review of optical coherence tomography angiography (OCTA). Int J Retina Vitreous 2015; 1: 5
• 24 Greig EC, Duker JS, Waheed NK. A practical guide to optical coherence tomography angiography interpretation. Int J Retina Vitreous 2020; 6: 55
• 25 Liu X, Huang ZY, Wang ZZ. et al. A deep learning based pipeline for optical coherence tomography angiography. J Biophotonics 2019; 12: e201900008
• 26 Jiang Z, Huang ZY, Qiu B. et al. Comparative study of deep learning models for optical coherence tomography angiography. Biomed Opt Express 2020; 11: 1580-1597
• 27 Lee CS, Tyring AJ, Wu Y. et al. Generating retinal flow maps from structural optical coherence tomography with artificial intelligence. Sci Rep 2019; 9: 5694
• 28 Li PL, OʼNeil C, Saberi S. et al. Deep learning algorithm for generating optical coherence tomography angiography (OCTA) maps of the retinal vasculature. In: Applications of Machine Learning 2020 Proc SPIE; 2020. 11511. 39-49
• 29 Chiu SJ, Li XT, Nicholas P. et al. Automatic segmentation of seven retinal layers in SDOCT images congruent with expert manual segmentation. Opt Express 2010; 18: 19413-19428
• 30 Antony BJ, Abràmoff MD, Lee K. et al. Automated 3D segmentation of intraretinal layers from optic nerve head optical coherence tomography images. In: Medical Imaging 2010: Biomedical Applications in Molecular, Structural, and Functional Imaging Proc SPIE; 2010. 7626. 249-260
• 31 Srinivasan PP, Heflin SJ, Izatt JA. et al. Automatic segmentation of up to ten layer boundaries in SD-OCT images of the mouse retina with and without missing layers due to pathology. Biomed Opt Express 2014; 5: 348-365
• 32 Vermeer KA, van der Schoot J, Lemij HG. et al. Automated segmentation by pixel classification of retinal layers in ophthalmic OCT images. Biomed Opt Express 2011; 2: 1743-1756
• 33 Dufour PA, Ceklic L, Abdillahi H. et al. Graph-based multi-surface segmentation of OCT data using trained hard and soft constraints. IEEE Trans Med Imaging 2013; 32: 531-543
• 34 Karri SPK, Chakraborthi D, Chatterjee J. Learning layer-specific edges for segmenting retinal layers with large deformations. Biomed Opt Express 2016; 7: 2888-2901
• 35 Fang LY, Cunefare D, Wang C. et al. Automatic segmentation of nine retinal layer boundaries in OCT images of non-exudative AMD patients using deep learning and graph search. Biomed Opt Express 2017; 8: 2732-2744
• 36 Mishra Z, Ganegoda A, Selicha J. et al. Automated Retinal Layer Segmentation Using Graph-based Algorithm Incorporating Deep-learning-derived Information. Sci Rep 2020; 10: 9541
• 37 Shah A, Zhou LX, Abramoff MD. et al. Multiple surface segmentation using convolution neural nets: application to retinal layer segmentation in OCT images. Biomed Opt Express 2018; 9: 4509-4526
• 38 Sedai S, Antony B, Rai R. et al. Uncertainty guided semi-supervised Segmentation of retinal Layers in OCT Images. In: International Conference on Medical Image Computing and Computer-Assisted Intervention – MICCAI 2019 Cham: Springer; 2019: 282-290
• 39 Schottenhamml J, Moult EM, Ploner SB. et al. OCT-OCTA segmentation: combining structural and blood flow information to segment Bruchʼs membrane. Biomed Opt Express 2021; 12: 84-99
• 40 Kadomoto S, Uji A, Muraoka Y. et al. Enhanced Visualization of Retinal Microvasculature in Optical Coherence Tomography Angiography Imaging via Deep Learning. J Clin Med 2020; 9: 1322
• 41 Niederleithner M, Britten A, Matten P. et al. 3D deep learning algorithm for denoising OCTA volumes acquired at 1.68 MHz A-scan-rate. Invest Ophthalmol Vis Sci 2021; 62: 65
• 42 Gao M, Guo YK, Hormel TT. et al. Reconstruction of high-resolution 6 × 6-mm OCT angiograms using deep learning. Biomed Opt Express 2020; 11: 3585-3600
• 43 Zhou T, Yang J, Zhou K. et al. Digital resolution enhancement in low transverse sampling optical coherence tomography angiography using deep learning. OSA Continuum 2020; 3: 1664-1678
• 44 Kim G, Kim J, Choi WJ. et al. Integrated deep learning framework for accelerated optical coherence tomography angiography. Sci Rep 2022; 12: 1289
• 45 Gao D, Celik N, Wu X. et al. A novel deep learning based OCTA de-striping method. In: Zheng Y, Williams BM, Chen K. eds. Medical Image Understanding and Analysis – 23rd Conference, MIUA 2019. Heidelberg: Springer; 2019: 189-197
• 46 Prentašic P, Heisler M, Mammo Z. et al. Segmentation of the foveal microvasculature using deep learning networks. J Biomed Opt 2016; 21: 75008
• 47 Lo J, Heisler M, Vanzan V. et al. Microvasculature Segmentation and Intercapillary Area Quantification of the Deep Vascular Complex Using Transfer Learning. Transl Vis Sci Technol 2020; 9: 38
• 48 Liu XX, Bi L, Xu YP. et al. Robust deep learning method for choroidal vessel segmentation on swept source optical coherence tomography images. Biomed Opt Express 2019; 10: 1601-1612
• 49 Li MC, Chen YR, Ji ZX. et al. Image Projection Network: 3D to 2D Image Segmentation in OCTA Images. IEEE Trans Med Imaging 2020; 39: 3343-3354
• 50 Alam M, Lim JI, Toslak D. et al. Differential Artery-Vein Analysis Improves the Performance of OCTA Staging of Sickle Cell Retinopathy. Transl Vis Sci Technol 2019; 8: 3
• 51 Alam M, Toslak D, Lim JI. et al. Color Fundus Image Guided Artery-Vein Differentiation in Optical Coherence Tomography Angiography. Invest Ophthalmol Vis Sci 2018; 59: 4953-4962
• 52 Son T, Alam M, Kim TH. et al. Near infrared oximetry-guided artery–vein classification in optical coherence tomography angiography. Exp Biol Med (Maywood) 2019; 244: 813-818
• 53 Alam M, Le D, Son T. et al. AV-Net: deep learning for fully automated artery-vein classification in optical coherence tomography angiography. Biomed Opt Express 2020; 11: 5249-5257
• 54 Woo JM, Yoon YS, Woo JE. et al. Foveal Avascular Zone Area Changes Analyzed Using OCT Angiography after Successful Rhegmatogenous Retinal Detachment Repair. Curr Eye Res 2018; 43: 674-678
• 55 Gill A, Cole ED, Novais EA. et al. Visualization of changes in the foveal avascular zone in both observed and treated diabetic macular edema using optical coherence tomography angiography. Int J Retina Vitreous 2017; 3: 17
• 56 Yoon YS, Woo JM, Woo JE. et al. Superficial foveal avascular zone area changes before and after idiopathic epiretinal membrane surgery. Int J Ophthalmol 2018; 11: 1711-1715
• 57 Balaratnasingam C, Inoue M, Ahn S. et al. Visual Acuity Is Correlated with the Area of the Foveal Avascular Zone in Diabetic Retinopathy and Retinal Vein Occlusion. Ophthalmology 2016; 123: 2352-2367
• 58 Guo M, Zhao M, Cheong AM. et al. Automatic quantification of superficial foveal avascular zone in optical coherence tomography angiography implemented with deep learning. Vis Comput Ind Biomed Art 2019; 2: 21
• 59 Guo ML, Zhao M, Cheong AMY. et al. Can deep learning improve the automatic segmentation of deep foveal avascular zone in optical coherence tomography angiography?. Biomed Signal Process Contr 2021; 66: 102456
• 60 Mirshahi R, Anvari P, Riazi-Esfahani H. et al. Foveal avascular zone segmentation in optical coherence tomography angiography images using a deep learning approach. Sci Rep 2021; 11: 1031
• 61 Lommatzsch C, Rothaus K, Koch M. et al. OCTA vessel density changes in the macular zone in glaucomatous eyes. Graefes Arch Clin Exp Ophthalmol 2018; 256: 1499-1508
• 62 Moghimi S, Zangwill LM, Penteado RC. et al. Macular and Optic Nerve Head Vessel Density and Progressive Retinal Nerve Fiber Layer Loss in Glaucoma. Ophthalmology 2018; 125: 1720-1728
• 63 Levine ES, Arya M, Chaudhari J. et al. Repeatability and reproducibility of vessel density measurements on optical coherence tomography angiography in diabetic retinopathy. Graefes Arch Clin Exp Ophthalmol 2020; 258: 1687-1695
• 64 Lavia C, Couturier A, Erginay A. et al. Reduced vessel density in the superficial and deep plexuses in diabetic retinopathy is associated with structural changes in corresponding retinal layers. PLos One 2019; 14: e0219164
• 65 Schottenhamml J, Moult EM, Ploner S. et al. An automatic, intercapillary area-based algorithm for quantifying diabetes-related capillary dropout using optical coherence tomography angiography. Retina 2016; 36: S93-S101
• 66 Lauermann P, van Oterendorp C, Storch MW. et al. Distance-Thresholded Intercapillary Area Analysis Versus Vessel-Based Approaches to Quantify Retinal Ischemia in OCTA. Transl Vis Sci Technol 2019; 8: 28
• 67 Sawada O, Ichiyama Y, Obata S. et al. Comparison between wide-angle OCT angiography and ultra-wide field fluorescein angiography for detecting non-perfusion areas and retinal neovascularization in eyes with diabetic retinopathy. Graefes Arch Clin Exp Ophthalmol 2018; 256: 1275-1280
• 68 Guo YK, Camino A, Wang J. et al. MEDnet, a neural network for automated detection of avascular area in OCT angiography. Biomed Opt Express 2018; 9: 5147-5158
• 69 Wang J, Hormel TT, You QS. et al. Robust non-perfusion area detection in three retinal plexuses using convolutional neural network in OCT angiography. Biomed Opt Express 2020; 11: 330-345
• 70 Wang J, Hormel TT, Gao LQ. et al. Automated diagnosis and segmentation of choroidal neovascularization in OCT angiography using deep learning. Biomed Opt Express 2020; 11: 927-944
• 71 Ong EP, Cheng J, Wong DW. et al. Glaucoma classification from retina optical coherence tomography angiogram. In: 2017 39th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC): IEEE; 2017: 596-599
• 72 Bowd C, Belghith A, Christopher M. et al. Deep-learning enface image classifier analysis of optical coherence tomography angiography images improves classification of healthy and glaucoma eyes. Invest Ophthalmol Vis Sci 2021; 62: 1024
• 73 Aslam TM, Hoyle DC, Puri V. et al. Differentiation of Diabetic Status Using Statistical and Machine Learning Techniques on Optical Coherence Tomography Angiography Images. Transl Vis Sci Technol 2020; 9: 2
• 74 Sandhu HS, Elmogy M, Sharafeldeen AT. et al. Automated Diagnosis of Diabetic Retinopathy Using Clinical Biomarkers, Optical Coherence Tomography, and Optical Coherence Tomography Angiography. Am J Ophthalmol 2020; 216: 201-206
• 75 Le D, Alam M, Yao CK. et al. Transfer Learning for Automated OCTA Detection of Diabetic Retinopathy. Transl Vis Sci Technol 2020; 9: 35
• 76 Heisler M, Karst S, Lo JL. et al. Ensemble Deep Learning for Diabetic Retinopathy Detection Using Optical Coherence Tomography Angiography. Transl Vis Sci Technol 2020; 9: 20
• 77 Zang PX, Gao LQ, Hormel TT. et al. DcardNet: Diabetic Retinopathy Classification at Multiple Levels Based on Structural and Angiographic Optical Coherence Tomography. IEEE Trans Biomed Eng 2021; 68: 1859-1870
• 78 Alfahaid A, Morris T. An Automated Age-Related Macular Degeneration Classification Based on Local Texture Features in Optical Coherence Tomography Angiography. In: Nixon M, Mahmoodi S, Zwiggelaar R. eds. Medical Image Understanding and Analysis – 22nd Conference, MIUA 2018. Heidelberg: Springer; 2018: 189-200
• 79 Alfahaid A, Morris T, Cootes T. et al. A Hybrid Machine Learning Approach Using LBP Descriptor and PCA for Age-Related Macular Degeneration Classification in OCTA Images. In: Zheng Y, Williams BM, Chen K. eds. Medical Image Understanding and Analysis – 23rd Conference, MIUA 2019. Heidelberg: Springer; 2019: 231-241
• 80 Jin K, Yan Y, Chen ML. et al. Multimodal deep learning with feature level fusion for identification of choroidal neovascularization activity in age-related macular degeneration. Acta Ophthalmol 2022; 100: E512-E520