Procedures of Establishing a Well-annotated Database of Color Fundus Photography of Diabetic Retinopathy for Artificial Intelligence Research
-
摘要: 近年来,随着人工智能技术的兴起,世界范围内对于眼底病尤其是糖尿病视网膜病变的人工智能识别研究越来越多,但不同人工智能识别软件准确度的验证与比较问题尚未解决。本文对北京协和医院眼科在建立1.5万张糖尿病视网膜病变眼底彩照人工智能研究标准数据库的过程中所遵循的规范进行阐述,以期为后续眼科以及其他相关学科建立更多的相关数据库提供参考。Abstract: With the development of artificial intelligence technology, explosive research on the recognition of retina diseases with artificial intelligence was published worldwide, among which, diabetic retinopathy attracted most attention. However, evaluation of different artificial intelligence software remains a problem. The Department of Ophthalmology, Peking Union Medical College Hospital published the first standard well-annotated color photo database of diabetic retinopathy for artificial intelligence research in China. We herein present the procedures for the establishment of the database and aim to provide reference for the future development of similar databases.
-
Key words:
- diabetic retinopathy /
- artificial intelligence /
- annotation /
- database /
- standard procedure
-
[1] 王亚星, 薛灿灿, 李建军. 眼科人工智能研究存在的主要问题及对策[J]. 眼科, 2021, 30: 81-84. https://www.cnki.com.cn/Article/CJFDTOTAL-YAKE202102004.htm [2] U.S. Food and Drug Administration. FDA permits marketing of artificial intelligence-based device to detect certain diabetes-related eye problems[EB/OL]. (2018-04-11)[2021-08-21]. https://www.fda.gov/news-events/press-announcements/fda-permits-marketing-artificial-intelligence-based-device-detect-certain-diabetes-related-eye. [3] Gulshan V, Peng L, Coram M, et al. Development and Validation of a Deep Learning Algorithm for Detection of Diabetic Retinopathy in Retinal Fundus Photographs[J]. JAMA, 2016, 316: 2402-2410. doi: 10.1001/jama.2016.17216 [4] 中华医学会眼科学会眼底病学组. 我国糖尿病视网膜病变临床诊疗指南(2014年)[J]. 中华眼科杂志, 2014, 50: 851-865. doi: 10.3760/cma.j.issn.0412-4081.2014.11.014 [5] Rivera SC, Liu X, Chan AW, et al. Guidelines for clinical trial protocols for interventions involving artificial intelligence: the SPIRIT-AI extension[J]. Nat Med, 2020, 26: 1351-1363. doi: 10.1038/s41591-020-1037-7