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

Biomed Opt Express. 2019 Jun 12;10(7):3257-3268. doi: 10.1364/BOE.10.003257. eCollection 2019 Jul 1.

Authors

Yukun Guo^{1

2}, Tristan T Hormel^{1

2}, Honglian Xiong^{1

3}, Bingjie Wang¹, Acner Camino¹, Jie Wang^{1

4}, David Huang¹, Thomas S Hwang¹, Yali Jia^{1

4}

Affiliations

¹ Casey Eye Institute, Oregon Health & Science University, Portland, OR 97239, USA.
² These authors contributed equally.
³ School of Physics and Optoelectronic Engineering, Foshan University, Foshan, Guangdong 528000, China.
⁴ Department of Biomedical Engineering, Oregon Health & Science University, Portland, OR 97239, USA.

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.

Abstract

Grants and funding