Advances on the application progress of artificial intelligence in diabetic retinopathy

doi:10.3877/cma.j.issn.2095-2007.2024.01.010

Abstract

Abstract:

Diabetic retinopathy (DR), as a common microvascular complication of diabetes, has become the main cause of blindness in young and middle age adults. However, the fundus screening for less than half of diabetes patients in China has performed. In recent years, researches on the artificial intelligence in the DR application field have increase gradually. Some artificial intelligence algorithms represented by deep learning through fundus photos, optical coherence tomography and other images recognition have revealed their advantages in the screening and grading of DR, which reduces the screening cost and lightens the burden of ophthalmologists. The application of artificial intelligence in DR from the aspects of risk prediction, screening and grading were reviewed in this paper, aiming to analyze the research situation, difficulties and development direction of artificial intelligence in the diagnosis and treatment of DR.

Key words: Artificial intelligence, Deep learning, Diabetic retinopathy, Risk prediction, Screening

Li Dong, Heyan Li, Wenbin Wei. Advances on the application progress of artificial intelligence in diabetic retinopathy[J]. Chinese Journal of Ophthalmologic Medicine(Electronic Edition), 2024, 14(01): 57-61.

/ / Recommend

Add to citation manager EndNote|Ris|BibTeX

URL: https://zhykyxzz.cma-cmc.com.cn/EN/10.3877/cma.j.issn.2095-2007.2024.01.010

https://zhykyxzz.cma-cmc.com.cn/EN/Y2024/V14/I01/57

Figures/Tables 1

References 47

[1]	万文萃，龙洋．糖尿病视网膜病变的流行病学、病因学与发病机制研究现状[J]. 眼科新进展，2022，42(9)：673－679.
[2]	Teo ZL, Tham YC, Yu M, et al. Global prevalence of diabetic retinopathy and projection of burden through 2045: systematic review and meta－analysis [J]. Ophthalmology, 2021, 128(11): 1580－1591.
[3]	Song P, Yu J, Chan KY, et al. Prevalence, risk factors and burden of diabetic retinopathy in China: A systematic review and meta－analysis[J]. J Glob Health, 2018, 8(1): e010803.
[4]	Peng J, Zou H, Wang W, et al. Implementation and first－year screening results of an ocular telehealth system for diabetic retinopathy in China[J]. BMC Health Serv Res, 2011, 11: 250.
[5]	Lundeen EA, Burke－Conte Z, Rein DB, et al. Prevalence of diabetic retinopathy in the US in 2021 [J]. JAMA Ophthalmol, 2023, 141(8): 747－754.
[6]	Liu G, Li Y, Pan A, et al. Adherence to a healthy lifestyle in association with microvascular complications among adults with type 2 diabetes [J]. JAMA Netw Open, 2023，6(1): e2252239.
[7]	Chen Z, Liu X, Hogan W, et al. Applications of artificial intelligence in drug development using real－world data[J]. Drug Discov Today, 2021, 26(5): 1256－1264.
[8]	Elsharkawy M, Sharafeldeen A, Khalifa F, et al. A clinically explainable AI－based grading system for age－related macular degeneration using optical coherence tomography[J]. IEEE J Biomed Health Inform, 2024, PMID: 38231804.
[9]	于楚瑶，董力，魏文斌．人工智能技术在病理性近视诊断与病情监测中的应用现状[J]. 中国医学前沿杂志(电子版)，2023，15(6)：15－20.
[10]	Wiedeman P，惠延年. 眼科学人工智能[J]. 国际眼科杂志，2023，23(9)：1417－1420.
[11]	魏文斌，李赫妍，董力. 重视数字眼科在眼科学发展中所发挥的重要作用[J]. 中国医学前沿杂志(电子版)，2023，15(6)：4－9，83.
[12]	Liu X, Faes L, Kale AU, et al. A comparison of deep learning performance against health－care professionals in detecting diseases from medical imaging: a systematic review and meta－analysis[J]. Lancet Digit Health, 2019, 1(6): e271－e297.
[13]	史雪辉，张丛，魏文斌．关注糖尿病黄斑水肿的光学相干断层扫描分型及相关影像特征[J]. 中华眼科医学杂志(电子版)，2021，11(1)：1－7.
[14]	Tsao HY, Chan PY, Su ECY. Predicting diabetic retinopathy and identifying interpretable biomedical features using machine learning algorithms[J]. BMC Bioinformatics, 2018, 19(9): 283.
[15]	王倩，延艳妮，杨婧研，等．不同程度糖尿病视网膜病变患者黄斑区视网膜血流变化及影响因素：开滦眼病研究[J]. 眼科，2022，31(4)：253－259.
[16]	Gardner GG, Keating D, Williamson TH, et al. Automatic detection of diabetic retinopathy using an artificial neural network: a screening tool[J]. Br J Ophthalmol, 1996, 80(11): 940－944.
[17]	Tang HL, Goh J, Peto T, et al. The reading of components of diabetic retinopathy: an evolutionary approach for filtering normal digital fundus imaging in screening and population based studies[J]. PLoS One, 2013, 8(7): e66730.
[18]	Ganesan K, Martis RJ, Acharya UR, et al. Computer－aided diabetic retinopathy detection using trace transforms on digital fundus images[J]. Med Biol Eng Comput, 2014, 52(8): 663－672.
[19]	毛明珠，黎梦宇，韦丁杨，等. 超广域SS－OCTA联合超广角激光扫描眼底成像对糖尿病视网膜病变的诊断价值[J]. 眼科新进展，2024，44(4)：291－296.
[20]	Lam C, Wong YL, Tang Z, et al. Performance of artificial intelligence in detecting diabetic macular edema from fundus photography and optical coherence tomography images: a systematic review and meta－analysis[J]. Diabetes Care, 2024, 47(2): 304－319.
[21]	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(22): 2402－2410.
[22]	Ting DSW, Cheung CYL, Lim G, et al. Development and validation of a deep learning system for diabetic retinopathy and related eye diseases using retinal images from multiethnic populations with diabetes[J]. JAMA, 2017, 318(22): 2211－2223.
[23]	Gargeya R, Leng T. Automated identification of diabetic retinopathy using deep learning[J]. Ophthalmology, 2017, 124(7): 962－969.
[24]	Sedova A, Hajdu D, Datlinger F, et al. Comparison of early diabetic retinopathy staging in asymptomatic patients between autonomous AI－based screening and human－graded ultra－widefield colour fundus images[J]. Eye (Lond), 2022, 6(3): 510－516.
[25]	Bhaskaranand M, Ramachandra C, Bhat S, et al. The value of automated diabetic retinopathy screening with the EyeArt system: A study of more than 100，000 consecutive encounters from people with diabetes[J]. Diabetes Technol Ther, 2019, 21(11): 635－643.
[26]	Zhang Y, Shi J, Peng Y, et al. Artificial intelligence－enabled screening for diabetic retinopathy: A real－world, multicenter and prospective study[J]. BMJ Open Diabetes Res Care, 2020, 8(1): e001596.
[27]	Keel S, Lee PY, Scheetz J, et al. Feasibility and patient acceptability of a novel artificial intelligence－based screening model for diabetic retinopathy at endocrinology outpatient services: a pilot study[J]. Sci Rep, 2018, 8(1): 4330.
[28]	Xie Y, Nguyen QD, Hamzah H, et al. Artificial intelligence for teleophthalmology－based diabetic retinopathy screening in a national programme: an economic analysis modelling study[J]. Lancet Digit Health, 2020, 2(5): e240－e249.
[29]	Hacisoftaoglu RE, Karakaya M, Sallam AB. Deep learning frameworks for diabetic retinopathy detection with smartphone－based retinal imaging systems[J]. Pattern Recognit Lett, 2020, 135: 409－417.
[30]	Sosale B, Aravind SR, Murthy H, et al. Simple, mobile－based artificial intelligence algorithm in the detection of diabetic retinopathy (SMART) study[J]. BMJ Open Diabetes Res Care, 2020, 8(1): e000892.
[31]	Abràmoff MD, Lavin PT, Birch M, et al. Pivotal trial of an autonomous AI－based diagnostic system for detection of diabetic retinopathy in primary care offices[J]. NPJ Digit Med, 2018, 1(1): 39.
[32]	Raumviboonsuk P, Krause J, Chotcomwongse P, et al. Deep learning versus human graders for classifying diabetic retinopathy severity in a nationwide screening program[J]. NPJ Digit Med, 2019, 2(1): 39.
[33]	Verbraak FD, Abramoff MD, Bausch GCF, et al. Diagnostic accuracy of a device for the automated detection of diabetic retinopathy in a primary care setting[J]. Diabetes Care, 2019, 42(4): 651－656.
[34]	He J, Cao T, Xu F, et al. Artificial intelligence－based screening for diabetic retinopathy at community hospital[J]. Eye, 2020, 34(3): 572－576.
[35]	Kanagasingam Y, Xiao D, Vignarajan J, et al. Evaluation of artificial intelligence－based grading of diabetic retinopathy in primary care[J]. JAMA Netw Open, 2018, 1(5): e182665.
[36]	Joseph S, Selvaraj J, Mani I, et al. Diagnostic accuracy of artificial intelligence－based automated diabetic retinopathy screening in real－world settings: a systematic review and meta－analysis[J]. Am J Ophthalmol, 2024, 263: 214－230.
[37]	Pei X, Yao X, Yang Y, et al. Efficacy of artificial intelligence－based screening for diabetic retinopathy in type 2 diabetes mellitus patients [J]. Diabetes Res Clin Pract, 2022, 184: 109190.
[38]	Li Z, Keel S, Liu C, et al. An automated grading system for detection of vision－threatening referable diabetic retinopathy on the basis of color fundus photographs[J]. Diabetes Care, 2018, 41(12): 2509－2516.
[39]	Ren F, Cao P, Zhao D, et al. Diabetic macular edema grading in retinal images using vector quantization and semi－supervised learning[J]. Technol Heal Care, 2018, 26(1): S389－S397.
[40]	Kermany DS, Goldbaum M, Cai W, et al. Identifying medical diagnoses and treatable diseases by image－based deep learning[J]. Cell, 2018, 172(5): 1122－1131.
[41]	Kim J, Tran L. Retinal disease classification from oct images using deep learning algorithms[C]// 2021 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB), 2021: 1－6.
[42]	Chalakkal R, Hafiz F, Abdulla W, et al. An efficient framework for automated screening of clinically significant macular edema[J]. Comput Biol Med, 2021, 130: 104128.
[43]	Cao J, You K, Jin K, et al. Prediction of response to anti－vascular endothelial growth factor treatment in diabetic macular oedema using an optical coherence tomography－based machine learning method[J]. Acta Ophthalmol, 2021, 99(1): e19－e27.
[44]	Paul W, Burlina P, Mocharla R, et al. Accuracy of artificial intelligence in estimating best－corrected visual acuity from fundus photographs in eyes with diabetic macular edema[J]. JAMA Ophthalmol, 2023, 141(7): 677－685.
[45]	Sabanayagam C, Xu D, Ting DSW, et al. A deep learning algorithm to detect chronic kidney disease from retinal photographs in community－based populations[J]. Lancet Digit Health, 2020, 2(6): e295－e302.
[46]	张娟娟，赵燕，许旺，等. 基于多指标的糖尿病患者视网膜病变预测评估模型研究[J]. 中国医院统计，2023，30(6)：401－407.
[47]	孙笑笑，陈有信，李东辉，等. 探究人工智能在眼底疾病筛查中的应用前景[J]. 中国眼镜科技杂志，2023，34(4)：118－122.

第一作者	年份	国家	人数	数据集			自主开发模型	模型	灵敏度(95%CI)	特异度(95%CI)
第一作者	年份	国家	人数	训练集	验证集	外部验证集	自主开发模型	模型	灵敏度(95%CI)	特异度(95%CI)
Gulshan等^[21]	2016	美国	75 444	EyePACS(128 175例)	EyePACS－1(9963例)	无	是	Inception－V3	90.1(87.2～92.6)	98.2(97.8～98.5)
					Messidor－2(1748例)				90.7(89.2～92.1)	93.8(93.2～94.4)
Daniel等^[22]	2017	新加坡	27 979	SIDRP(76 370例)	SIDRP(71 896例)	10	是	VGGNet	90.5(87.3～93.0)	91.6(91.0～92.2)
Gargeya等^[23]	2017	美国	无	EyePACS(75 137例)	15 000例	2	是	CNN	94.0	98.0
Sedova等^[24]	2022	奥地利	54	无	无	无	否	IDx－DR	100.0	47.0
Bhaskaranand等^[25]^*	2019	美国	101 710	无	无	EyePACS(综合医院850 908例)	否	EyeArt	91.3(90.9～91.7)	91.1(90.9～91.3)
Zhang等^[26]	2020	中国	47 269	无	无	(143 626例)公开数据库(1184例)	否	VoxelCloud Retina	83.3(81.9～84.6)	92.5(92.1～92.9)
Hacisoftaoglu等^[29]	2020	美国	无	无	无	UoA－DR(200例)	否	ResNet50	98.2	99.1
Sosale等^[30]	2019	印度	900	无	无	EyePACS(34 278例) Remidio FOP(4350例)	否	Medios AI	93.0(91.3～94.7)	92.5(90.8～94.2)
Abràmoff等^[31]	2018	美国	819	无	无	基层医院(1638例)	否	IDx－DR	87.2(81.8～91.2)	90.7(88.3～92.7)
Raumviboonsuk等^[32]	2019	泰国	7517	无	无	国家筛查(25 326例)	否	Inception－V4	96.8(89.3～99.3)	95.6(98.3～98.7)
Verbraak等^[33]*	2019	荷兰	1293	无	无	综合医院(2586例)	否	IDx－DR	79.4(66.5～87.9)	93.8(92.1～94.9)
He等^[34]^*	2019	中国	889	无	无	基层医院(3556例)	否	Inception－V4	91.2(86.4～94.7)	98.8(98.1～99.3)
Kanagasingam等^[35]	2018	澳大利亚	192	无	无	DiaRetDBI、EyePACS(30 000例)	否	Inception－V3	100.0(19.8～100.0)	92.1(87.1～95.4)

第一作者	年份	国家	人数	数据集			自主开发模型	模型	灵敏度(95%CI)	特异度(95%CI)
第一作者	年份	国家	人数	训练集	验证集	外部验证集	自主开发模型	模型	灵敏度(95%CI)	特异度(95%CI)
Gulshan等^[21]	2016	美国	75 444	EyePACS(128 175例)	EyePACS－1(9963例)	无	是	Inception－V3	90.1(87.2～92.6)	98.2(97.8～98.5)
					Messidor－2(1748例)				90.7(89.2～92.1)	93.8(93.2～94.4)
Daniel等^[22]	2017	新加坡	27 979	SIDRP(76 370例)	SIDRP(71 896例)	10	是	VGGNet	90.5(87.3～93.0)	91.6(91.0～92.2)
Gargeya等^[23]	2017	美国	无	EyePACS(75 137例)	15 000例	2	是	CNN	94.0	98.0
Sedova等^[24]	2022	奥地利	54	无	无	无	否	IDx－DR	100.0	47.0
Bhaskaranand等^[25]^*	2019	美国	101 710	无	无	EyePACS(综合医院850 908例)	否	EyeArt	91.3(90.9～91.7)	91.1(90.9～91.3)
Zhang等^[26]	2020	中国	47 269	无	无	(143 626例)公开数据库(1184例)	否	VoxelCloud Retina	83.3(81.9～84.6)	92.5(92.1～92.9)
Hacisoftaoglu等^[29]	2020	美国	无	无	无	UoA－DR(200例)	否	ResNet50	98.2	99.1
Sosale等^[30]	2019	印度	900	无	无	EyePACS(34 278例) Remidio FOP(4350例)	否	Medios AI	93.0(91.3～94.7)	92.5(90.8～94.2)
Abràmoff等^[31]	2018	美国	819	无	无	基层医院(1638例)	否	IDx－DR	87.2(81.8～91.2)	90.7(88.3～92.7)
Raumviboonsuk等^[32]	2019	泰国	7517	无	无	国家筛查(25 326例)	否	Inception－V4	96.8(89.3～99.3)	95.6(98.3～98.7)
Verbraak等^[33]*	2019	荷兰	1293	无	无	综合医院(2586例)	否	IDx－DR	79.4(66.5～87.9)	93.8(92.1～94.9)
He等^[34]^*	2019	中国	889	无	无	基层医院(3556例)	否	Inception－V4	91.2(86.4～94.7)	98.8(98.1～99.3)
Kanagasingam等^[35]	2018	澳大利亚	192	无	无	DiaRetDBI、EyePACS(30 000例)	否	Inception－V3	100.0(19.8～100.0)	92.1(87.1～95.4)

Please choose a citation manager

Content to export