155 262

Cited 24 times in

Detection of features associated with neovascular age-related macular degeneration in ethnically distinct data sets by an optical coherence tomography: trained deep learning algorithm

Authors
 Tyler Hyungtaek Rim  ;  Aaron Y Lee  ;  Daniel S Ting  ;  Kelvin Teo  ;  Bjorn Kaijun Betzler  ;  Zhen Ling Teo  ;  Tea Keun Yoo  ;  Geunyoung Lee  ;  Youngnam Kim  ;  Andrew C Lin  ;  Seong Eun Kim  ;  Yih Chung Tham  ;  Sung Soo Kim  ;  Ching-Yu Cheng  ;  Tien Yin Wong  ;  Chui Ming Gemmy Cheung 
Citation
 BRITISH JOURNAL OF OPHTHALMOLOGY, Vol.105(8) : 1133-1139, 2021-08 
Journal Title
BRITISH JOURNAL OF OPHTHALMOLOGY
ISSN
 0007-1161 
Issue Date
2021-08
MeSH
Aged ; Algorithms ; Area Under Curve ; Asians / ethnology* ; Choroidal Neovascularization / diagnostic imaging* ; Choroidal Neovascularization / ethnology ; Datasets as Topic ; Deep Learning ; Female ; Humans ; Image Interpretation, Computer-Assisted* ; Male ; Middle Aged ; Republic of Korea / epidemiology ; Tomography, Optical Coherence* ; Wet Macular Degeneration / diagnostic imaging* ; Wet Macular Degeneration / ethnology
Keywords
Degeneration ; Epidemiology ; Neovascularisation ; Retina.
Abstract
Background: The ability of deep learning (DL) algorithms to identify eyes with neovascular age-related macular degeneration (nAMD) from optical coherence tomography (OCT) scans has been previously established. We herewith evaluate the ability of a DL model, showing excellent performance on a Korean data set, to generalse onto an American data set despite ethnic differences. In addition, expert graders were surveyed to verify if the DL model was appropriately identifying lesions indicative of nAMD on the OCT scans.

Methods: Model development data set-12 247 OCT scans from South Korea; external validation data set-91 509 OCT scans from Washington, USA. In both data sets, normal eyes or eyes with nAMD were included. After internal testing, the algorithm was sent to the University of Washington, USA, for external validation. Area under the receiver operating characteristic curve (AUC) and precision-recall curve (AUPRC) were calculated. For model explanation, saliency maps were generated using Guided GradCAM.

Results: On external validation, AUC and AUPRC remained high at 0.952 (95% CI 0.942 to 0.962) and 0.891 (95% CI 0.875 to 0.908) at the individual level. Saliency maps showed that in normal OCT scans, the fovea was the main area of interest; in nAMD OCT scans, the appropriate pathological features were areas of model interest. Survey of 10 retina specialists confirmed this.

Conclusion: Our DL algorithm exhibited high performance for nAMD identification in a Korean population, and generalised well to an ethnically distinct, American population. The model correctly focused on the differences within the macular area to extract features associated with nAMD.
Files in This Item:
T202124872.pdf Download
DOI
10.1136/bjophthalmol-2020-316984
Appears in Collections:
1. College of Medicine (의과대학) > Dept. of Ophthalmology (안과학교실) > 1. Journal Papers
Yonsei Authors
Kim, Sung Soo(김성수) ORCID logo https://orcid.org/0000-0002-0574-7993
URI
https://ir.ymlib.yonsei.ac.kr/handle/22282913/187569
사서에게 알리기
  feedback

qrcode

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

Browse

Links