Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Review
. 2023 Jan 16;13(2):326.
doi: 10.3390/diagnostics13020326.

Deep Learning in Optical Coherence Tomography Angiography: Current Progress, Challenges, and Future Directions

Dawei Yang  1 , An Ran Ran  1 , Truong X Nguyen  1 , Timothy P H Lin  1   2 , Hao Chen  3 , Timothy Y Y Lai  1   4 , Clement C Tham  1   2 , Carol Y Cheung  1
Affiliations
Review

Deep Learning in Optical Coherence Tomography Angiography: Current Progress, Challenges, and Future Directions

Dawei Yang et al. Diagnostics (Basel). .

Abstract

Optical coherence tomography angiography (OCT-A) provides depth-resolved visualization of the retinal microvasculature without intravenous dye injection. It facilitates investigations of various retinal vascular diseases and glaucoma by assessment of qualitative and quantitative microvascular changes in the different retinal layers and radial peripapillary layer non-invasively, individually, and efficiently. Deep learning (DL), a subset of artificial intelligence (AI) based on deep neural networks, has been applied in OCT-A image analysis in recent years and achieved good performance for different tasks, such as image quality control, segmentation, and classification. DL technologies have further facilitated the potential implementation of OCT-A in eye clinics in an automated and efficient manner and enhanced its clinical values for detecting and evaluating various vascular retinopathies. Nevertheless, the deployment of this combination in real-world clinics is still in the "proof-of-concept" stage due to several limitations, such as small training sample size, lack of standardized data preprocessing, insufficient testing in external datasets, and absence of standardized results interpretation. In this review, we introduce the existing applications of DL in OCT-A, summarize the potential challenges of the clinical deployment, and discuss future research directions.

Keywords: artificial intelligence; deep learning; diabetic macular ischemia; diabetic retinopathy; glaucoma; image quality; medical image analysis; optical coherence tomography angiography; retinal vascular diseases.

PubMed Disclaimer

Conflict of interest statement

No conflicting relationship exists for any author.

Figures

Figure 1
Figure 1
Examples of a series of 3 × 3 mm2 superficial capillary plexus (SCP) and deep capillary plexus (DCP) optical coherence tomography angiography (OCT-A) images illustrating different severities of diabetic retinopathy. (A1–E1): SCP OCT-A images illustrating the alteration of the FAZ area and the surrounding vasculature from no DR to PDR. (A2–E2): DCP OCT-A images illustrating the alteration of the FAZ area and the surrounding vasculature from no DR to PDR. OCT-A: optical coherence tomography angiography; SCP: superficial capillary plexus; DCP: deep capillary plexus; FAZ: foveal avascular zone; DR: diabetic retinopathy; PDR: proliferate diabetic retinopathy.
Figure 2
Figure 2
Illustration of different kinds of artifacts in superficial capillary plexus (SCP) and deep capillary plexus (DCP) optical coherence tomography angiography (OCT-A) images. (A–C) Artifacts in SCP OCT-A images, namely, (A) movement artifact (red arrow); (B) defocus artifact; (C) shadow artifact. (D–F) Artifacts in DCP OCT-A images, namely, (D) projection artifact (red arrow); (E) defocus artifact; (F) shadow artifact.
Figure 3
Figure 3
An example illustrating that the 6 × 6 mm2 optical coherence tomography angiography (OCT-A) scanning protocol with larger field of view can detect non-perfusion area (red arrow) outside of the standard centralized 3 × 3 mm2 scanning protocol.
Figure 4
Figure 4
Examples of optical coherence tomography angiography (OCT-A) images in classifications of diabetic macular ischemia (DMI) on superficial capillary plexus (SCP) and deep capillary plexus (DCP). DMI classification from left to right: no DMI on SCP (A), DMI on SCP (B), no DMI on DCP (C), DMI on DCP (D). OCT-A: optical coherence tomography angiography; SCP: superficial capillary plexus; DCP: deep capillary plexus; DMI: diabetic macular ischemia.
See this image and copyright information in PMC

References

    1. Spaide R.F., Fujimoto J.G., Waheed N.K., Sadda S.R., Staurenghi G. Optical coherence tomography angiography. Prog. Retin. Eye Res. 2018;64:1–55. doi: 10.1016/j.preteyeres.2017.11.003. - DOI - PMC - PubMed
    1. Sun Z., Yang D., Tang Z., Ng D.S., Cheung C.Y. Optical coherence tomography angiography in diabetic retinopathy: An updated review. Eye. 2021;35:149–161. doi: 10.1038/s41433-020-01233-y. - DOI - PMC - PubMed
    1. Cavichini M., Dans K.C., Jhingan M., Amador-Patarroyo M.J., Borooah S., Bartsch D.U., Nudleman E., Freeman W.R. Evaluation of the clinical utility of optical coherence tomography angiography in age-related macular degeneration. Br. J. Ophthalmol. 2021;105:983–988. doi: 10.1136/bjophthalmol-2020-316622. - DOI - PubMed
    1. Wang Y.M., Shen R., Lin T.P.H., Chan P.P., Wong M.O.M., Chan N.C.Y., Tang F., Lam A.K.N., Leung D.Y.L., Tham C.C.Y., et al. Optical coherence tomography angiography metrics predict normal tension glaucoma progression. Acta Ophthalmol. 2022;100:e1455–e1462. doi: 10.1111/aos.15117. - DOI - PubMed
    1. Yeung L., Wu W.C., Chuang L.H., Wang N.K., Lai C.C. Novel Optical Coherence Tomography Angiography Biomarker in Branch Retinal Vein Occlusion Macular Edema. Retina. 2019;39:1906–1916. doi: 10.1097/IAE.0000000000002264. - DOI - PubMed

LinkOut - more resources