Development and validation of a deep learning algorithm for distinguishing the nonperfusion area from signal reduction artifacts on OCT angiography

Yukun Guo; Tristan T. Hormel; Honglian Xiong; Bingjie Wang; Acner Camino; Jie Wang; David Huang; Thomas S. Hwang; Yali Jia

doi:10.1364/BOE.10.003257

Development and validation of a deep learning algorithm for distinguishing the nonperfusion area from signal reduction artifacts on OCT angiography

Yukun Guo, Tristan T. Hormel, Honglian Xiong, Bingjie Wang, Acner Camino, Jie Wang, David Huang, Thomas S. Hwang, Yali Jia

Ophthalmology

Research output: Contribution to journal › Article › peer-review

41 Scopus citations

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 mm2 OCTA. The network achieves strong specificity and sensitivity for NPA detection across a wide range of DR severity and scan quality.

Original language	English (US)
Pages (from-to)	3257-3268
Number of pages	12
Journal	Biomedical Optics Express
Volume	10
Issue number	7
DOIs	https://doi.org/10.1364/BOE.10.003257
State	Published - Jul 2019

ASJC Scopus subject areas

Biotechnology
Atomic and Molecular Physics, and Optics

Access to Document

10.1364/BOE.10.003257

Cite this

@article{77ded75fce554214861751f616b28d35,

title = "Development and validation of a deep learning algorithm for distinguishing the nonperfusion area from signal reduction artifacts on OCT angiography",

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 mm2 OCTA. The network achieves strong specificity and sensitivity for NPA detection across a wide range of DR severity and scan quality.",

author = "Yukun Guo and Hormel, {Tristan T.} and Honglian Xiong and Bingjie Wang and Acner Camino and Jie Wang and David Huang and Hwang, {Thomas S.} and Yali Jia",

note = "Publisher Copyright: {\textcopyright} 2019 Optical Society of America under the terms of the OSA Open Access Publishing Agreement.",

year = "2019",

month = jul,

doi = "10.1364/BOE.10.003257",

language = "English (US)",

volume = "10",

pages = "3257--3268",

journal = "Biomedical Optics Express",

issn = "2156-7085",

publisher = "The Optical Society",

number = "7",

}

TY - JOUR

T1 - Development and validation of a deep learning algorithm for distinguishing the nonperfusion area from signal reduction artifacts on OCT angiography

AU - Guo, Yukun

AU - Hormel, Tristan T.

AU - Xiong, Honglian

AU - Wang, Bingjie

AU - Camino, Acner

AU - Wang, Jie

AU - Huang, David

AU - Hwang, Thomas S.

AU - Jia, Yali

PY - 2019/7

Y1 - 2019/7

N2 - 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 mm2 OCTA. The network achieves strong specificity and sensitivity for NPA detection across a wide range of DR severity and scan quality.

AB - 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 mm2 OCTA. The network achieves strong specificity and sensitivity for NPA detection across a wide range of DR severity and scan quality.

UR - http://www.scopus.com/inward/record.url?scp=85070749019&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85070749019&partnerID=8YFLogxK

U2 - 10.1364/BOE.10.003257

DO - 10.1364/BOE.10.003257

M3 - Article

AN - SCOPUS:85070749019

SN - 2156-7085

VL - 10

SP - 3257

EP - 3268

JO - Biomedical Optics Express

JF - Biomedical Optics Express

IS - 7

ER -

Development and validation of a deep learning algorithm for distinguishing the nonperfusion area from signal reduction artifacts on OCT angiography

Abstract

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this