Affordable Access

deepdyve-link
Publisher Website

An Artificial-Intelligence-Based Automated Grading and Lesions Segmentation System for Myopic Maculopathy Based on Color Fundus Photographs.

Authors
  • Tang, Jia1, 2
  • Yuan, Mingzhen1, 2
  • Tian, Kaibin3
  • Wang, Yuelin1, 2
  • Wang, Dongyue1, 2
  • Yang, Jingyuan1, 2
  • Yang, Zhikun1
  • He, Xixi4
  • Luo, Yan1
  • Li, Ying1
  • Xu, Jie5
  • Li, Xirong3, 6
  • Ding, Dayong4
  • Ren, Yanhan7
  • Chen, Youxin1, 2
  • Sadda, Srinivas R8, 9
  • Yu, Weihong1, 2
  • 1 Department of Ophthalmology, Peking Union Medical College Hospital, Beijing, China. , (China)
  • 2 Key Laboratory of Ocular Fundus Diseases, Chinese Academy of Medical Sciences, Beijing, China. , (China)
  • 3 AI and Media Computing Lab, School of Information, Renmin University of China, Beijing, China. , (China)
  • 4 Vistel AI Lab, Visionary Intelligence, Beijing, China. , (China)
  • 5 Beijing Institute of Ophthalmology, Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical University, Beijing Ophthalmology and Visual Science Key Laboratory, Beijing, China. , (China)
  • 6 Key Laboratory of Data Engineering and Knowledge Engineering, Renmin University of China, Beijing, China. , (China)
  • 7 Chicago Medical School, Rosalind Franklin University of Medicine and Science, North Chicago, IL, USA.
  • 8 Doheny Eye Institute, Los Angeles, CA, USA.
  • 9 Department of Ophthalmology, University of California, Los Angeles, Los Angeles, CA, USA.
Type
Published Article
Journal
Translational Vision Science & Technology
Publisher
Association for Research in Vision and Ophthalmology (ARVO)
Publication Date
Jun 01, 2022
Volume
11
Issue
6
Pages
16–16
Identifiers
DOI: 10.1167/tvst.11.6.16
PMID: 35704327
Source
Medline
Language
English
License
Unknown

Abstract

To develop deep learning models based on color fundus photographs that can automatically grade myopic maculopathy, diagnose pathologic myopia, and identify and segment myopia-related lesions. Photographs were graded and annotated by four ophthalmologists and were then divided into a high-consistency subgroup or a low-consistency subgroup according to the consistency between the results of the graders. ResNet-50 network was used to develop the classification model, and DeepLabv3+ network was used to develop the segmentation model for lesion identification. The two models were then combined to develop the classification-and-segmentation-based co-decision model. This study included 1395 color fundus photographs from 895 patients. The grading accuracy of the co-decision model was 0.9370, and the quadratic-weighted κ coefficient was 0.9651; the co-decision model achieved an area under the receiver operating characteristic curve of 0.9980 in diagnosing pathologic myopia. The photograph-level F1 values of the segmentation model identifying optic disc, peripapillary atrophy, diffuse atrophy, patchy atrophy, and macular atrophy were all >0.95; the pixel-level F1 values for segmenting optic disc and peripapillary atrophy were both >0.9; the pixel-level F1 values for segmenting diffuse atrophy, patchy atrophy, and macular atrophy were all >0.8; and the photograph-level recall/sensitivity for detecting lacquer cracks was 0.9230. The models could accurately and automatically grade myopic maculopathy, diagnose pathologic myopia, and identify and monitor progression of the lesions. The models can potentially help with the diagnosis, screening, and follow-up for pathologic myopic in clinical practice.

Report this publication

Statistics

Seen <100 times