[1]郑武,阮坤炜,吴天添,等.人工智能糖尿病视网膜病变筛查系统与眼科医师诊断结果的一致性分析[J].眼科新进展,2020,40(12):1170-1173.[doi:10.13389/j.cnki.rao.2020.0260]
　ZHENG Wu,RUAN Kunwei,WU Tiantian,et al.Consistency of artificial intelligence screening system with ophthalmologist for diagnosing of diabetic retinopathy[J].Recent Advances in Ophthalmology,2020,40(12):1170-1173.[doi:10.13389/j.cnki.rao.2020.0260]
点击复制

人工智能糖尿病视网膜病变筛查系统与眼科医师诊断结果的一致性分析/HTML
分享到：

《眼科新进展》[ISSN:1003-5141/CN:41-1105/R]

卷:
40卷
期数:
2020年12期
页码:
1170-1173
栏目:
应用研究
出版日期:
2020-12-05

文章信息/Info

Title:
Consistency of artificial intelligence screening system with ophthalmologist for diagnosing of diabetic retinopathy
文章编号:
1170
作者:
郑武; 阮坤炜; 吴天添; 瞿婕; 李欣; 张勇
442000　湖北省十堰市，湖北医药学院（郑武，吴天添）；442000　湖北省十堰市，十堰市太和医院眼科中心（阮坤炜，李欣，张勇）；121000　辽宁省锦州市，锦州医科大学（瞿婕）
Author(s):
ZHENG Wu1; RUAN Kunwei2; WU Tiantian1; QU Jie3; LI Xin2; ZHANG Yong2
1.Hubei University of Medicine,Shiyan 442000,Hubei Province,China
2.Department of Ophthalmology，Taihe Hospital of Shiyan,Shiyan 442000,Hubei Province,China
3.Jinzhou Medical University,Jinzhou 121000,Liaoning Province,China
关键词:
糖尿病视网膜病变; 筛查; 眼底彩色照相; 人工智能; 一致性分析
Keywords:
diabetic retinopathy;  screening;  fundus color photography;  artificial intelligence;  consistency analysis
分类号:
R774
DOI:
10.13389/j.cnki.rao.2020.0260
文献标志码:
A
摘要:
目的　探讨人工智能糖尿病视网膜病变（diabetic retinopathy,DR）筛查系统与临床医师诊断的一致性。方法　收集2019年3月至2020年5月在十堰市太和医院眼科就诊的糖尿病患者325例（650眼）的眼科影像学检查资料，包括眼底彩色照相、黄斑区OCT、荧光素眼底血管造影（FFA），由专家组和人工智能DR筛查系统分别给出诊断，专家组参考眼底彩色照相、黄斑区OCT及FFA检查结果做出诊断,人工智能DR筛查系统仅以眼底彩色照相结果作为诊断依据。以专家组诊断结果为标准，分析人工智能DR筛查系统诊断各级DR的敏感性、特异度、漏诊率和误诊率，对二者诊断结果的一致性行Kappa检验分析。结果　与专家组诊断结果相比,人工智能诊断有无DR的敏感性为94.78%，特异度为92.11%,二者诊断结果一致性分析的Kappa值为0.82（P<0.001)。与专家组诊断结果相比，人工智能诊断有转诊意义的DR的敏感性为93.75%，特异度为96.18%,二者对DR分期诊断的一致性分析的Kappa值为0.85（P<0.001)。结论　人工智能DR筛查系统与专家组诊断结果一致性高，具有较高的准确性和敏感度，可大规模用于DR的初级筛查工作。
Abstract:
Objective　To explore the consistency of artificial intelligence (AI) screening system for diabetic retinopathy (DR)with clinicians.Methods　The ophthalmic imaging data of 325 diabetic patients (612 eyes) in the Department of Ophthalmology of Taihe Hospital of Shiyan City from March 2019 to May 2020 were collected, including fundus color photo, macular OCT and fundus fluorescein angiography (FFA), and were diagnosed by expert group and AI group, respectively. The expert group referred to fundus photography, macular OCT and FFA, AI group only based on fundus color photography. According to the diagnostic results of expert group, the sensitivity, specificity, missed diagnosis rate and misdiagnosis rate of artificial intelligence DR screening system and the consistency of Kappa test were analyzed.Results　When compared with expert group, the Kappa value of AI diagnosis with or without DR was 0.82, the sensitivity was 94.78%, and the specificity was 92.11% (P<0.001). Compared with the diagnosis results of the expert group, the sensitivity of AI group with referral significance was 93.75%, and the specificity was 96.18%. The Kappa value of the consistency analysis of the two diagnosis of DR staging was 0.85 (P<0.001).Conclusion　The AI screening DR system has high accuracy and sensitivity, and has high consistency with the diagnosis of the expert group, so it can be widely used in the primary screening of DR.

参考文献/References:

［1］　葛景利,代青湘.糖尿病肾病治疗的研究进展综述［J］.外科,2019,8(3):78-83.
GE J L,DAI Q X.Progress in the treatment of diabetic nephropathy［J］.Hans J Surg,2019,8(3):78-83.
［2］　YAU J W,ROGERS S L,KAWASAKI R,LAMOUREUX E L,KOWALSKI J W,TOKE B,et al.Global prevalence and major risk factors of diabetic retinopathy［J］.Diabetes Care,2012,35(3):556-564.
［3］　LIU L,WU X M,LIU L M,GENG J,YUAN Z,SHAN Z Y,et al.Prevalence of diabetic retinopathy in mainland China: a meta-analysis［J］.PLoS One,2012,7(9):e45264.
［4］　邓启凤,陈静,邹玉平,张楚，许晓珊.2型糖尿病患者合并糖尿病视网膜病变的危险因素分析［J］.海军医学杂志,2020,41(2):162-164.
DENG Q F,CHEN J,ZOU Y P,ZHANG C,XU X S.Analysis of risk factors of type 2 diabetes mellitus patients complicated with diabetic retinopathy［J］.J Navy Med,2020,41(2):162-164.
［5］　中华医学会糖尿病学分会视网膜病变学组.糖尿病视网膜病变防治专家共识［J］.中华糖尿病杂志,2018,10(4):241-247.
Retinopathy Group of Diabetes Branch of Chinese Medical Association.Consensus of experts on prevention and treatment of diabetic retinopathy［J］.Chin J Diabetes Mellitus,2018,10(4):241-247.
［6］　王莎莎.人工智能在眼部图像诊断中的应用［J］.中华实验眼科杂志,2018,36(10):796-799.
WANG S S，Application of artificial intelligence in ocular imaging diagnosis［J］.Chin J Exp Ophthalmol,2018,36(10):796-799.
［7］　ABRMOFF M D,FOLK J C,HAN D P,WALKER J D,WILLIAMS D F,RUSSELL S R,et al.Automated analysis of retinal images for detection of referable diabetic retinopathy［J］.JAMA Ophthalmol,2013,131(3):351-357.
［8］　TING D,CHEUNG C Y,LIM G,TAN G,QUANG N,GAN A,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］.J Am Med Assoc,2017,318(22):2211-2223.
［9］　TING D,PENG L,VARADARAJAN A V,KEANE P A,BURLINA P M,CHIANG M F,et al.Deep learning in ophthalmology: The technical and clinical considerations［J］.Prog Retin Eye Res,2019,72:100759.
［10］　BHASKARANAND M,RAMACHANDRA C,BHAT S,CUADROS J,NITTALA M G,SADDA S R,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.
［11］　ABRAMOFF M D,LAVIN P T,BIRCH M,SHAH N,FOLK J C.Pivotal trial of an autonomous AI-based diagnostic system for detection of diabetic retinopathy in primary care offices［J］.NPJ Digit Med,2018,1:39.
［12］　RUAMVIBOONSUK P,CHEUNG C Y,ZHANG X,RAMAN R,PARK S J,TING D,et al.Artificial intelligence in ophthalmology: Evolutions in Asia［J］.Asia Pac J Ophthalmol (Phila),2020,9(2):78-84.
［13］　王雯秋,孙晓东.人工智能辅助眼底影像 ——机遇与挑战并存［J］.中华眼视光学与视觉科学杂志,2019,21(3):166-169.
WANG W Q,SUN X D.Artificial intelligence analysis of fundus images—Opportunities and challenges［J］.Chin J Optom Ophthalmol Vis Sci,2019,21(3):166-169.

相似文献/References:

[1]杨倩 韩毳 刘宁 刘希娥 徐青 侯世科.免散瞳数码眼底照相在眼底黄斑部疾病筛查中的应用[J].眼科新进展,2012,32(5):000.
[2]杜玮 刘子扬 周艳艳 雒雷鸣.糖尿病视网膜病变与血清胆红素水平的关系[J].眼科新进展,2012,32(5):000.
[3]范松涛 卢建民.阿司匹林与糖尿病患者玻璃体出血以及玻璃体切割术疗效的相关性研究[J].眼科新进展,2012,32(11):000.
[4]李艳 李筱荣 袁佳琴 潘斌.糖尿病大鼠视网膜中VEGF、PEDF的表达与血-视网膜屏障损伤[J].眼科新进展,2013,33(1):000.
[5]李朝晖 崔治华 胡晓英 孟丽珠 张敬维.糖尿病视网膜病变激光面积与疗效的分析[J].眼科新进展,2013,33(2):000.
[6]冯冬梅 朱鸿 施彩虹.CXC趋化因子及其受体在糖尿病视网膜病变中的作用[J].眼科新进展,2013,33(6):000.
[7]牛淑玲.糖尿病视网膜病变患者HbAlc、FPG与血小板参数的变化及危险因素分析[J].眼科新进展,2013,33(7):000.
[8]毕春潮 王睿 王建洲 雷春灵 董晓娟 王小莉 薛晓辉.Ad-PEDF对糖尿病视网膜病变大鼠视网膜新生血管的抑制作用[J].眼科新进展,2013,33(8):000.
[9]杨萍 孙书明 李晓鹏.辛伐他汀对糖尿病视网膜病变和炎症因子的影响[J].眼科新进展,2013,33(8):000.
[10]罗文婷 孙大卫.血管黏附蛋白-1在眼科疾病中的研究进展[J].眼科新进展,2013,33(8):000.

备注/Memo

备注/Memo:
十堰市科技局基金资助（编号：18Y32）；太和医院横向课题基金资助（编号：2017TJXM001）
更新日期/Last Update: 2020-12-05