An Evaluation System of Fundus Photograph-Based Intelligent Diagnostic Technology for Diabetic Retinopathy and Applicability for Research

被引:36
|
作者
Yang, Wei-Hua [1 ,2 ,4 ]
Zheng, Bo [3 ,4 ]
Wu, Mao-Nian [3 ,4 ]
Zhu, Shao-Jun [3 ,4 ]
Fei, Fang-Qin [4 ,7 ]
Weng, Ming [5 ]
Zhang, Xian [6 ]
Lu, Pei-Rong [1 ]
机构
[1] Soochow Univ, Affiliated Hosp 1, Dept Ophthalmol, Suzhou, Jiangsu, Peoples R China
[2] First Peoples Hosp Huzhou, Dept Ophthalmol, Huzhou, Zhejiang, Peoples R China
[3] Huzhou Univ, Informat Engn Coll, Huzhou, Zhejiang, Peoples R China
[4] Huzhou Univ, Key Lab Med Artificial Intelligence, Huzhou, Zhejiang, Peoples R China
[5] Wuxi Third Peoples Hosp, Dept Ophthalmol, Wuxi, Jiangsu, Peoples R China
[6] Lihuili Eastern Hosp, Ningbo Med Ctr, Dept Ophthalmol, Ningbo, Zhejiang, Peoples R China
[7] Huzhou Univ, Affiliated Hosp 1, Dept Endocrinol, Huzhou, Zhejiang, Peoples R China
关键词
Deep learning; Diabetic retinopathy; Evaluation studies; Ophthalmological diagnostic techniques; DEEP LEARNING ALGORITHM; VESSEL SEGMENTATION;
D O I
10.1007/s13300-019-0652-0
中图分类号
R5 [内科学];
学科分类号
1002 ; 100201 ;
摘要
Introduction In April 2018, the US Food and Drug Administration (FDA) approved the world's first artificial intelligence (AI) medical device for detecting diabetic retinopathy (DR), the IDx-DR. However, there is a lack of evaluation systems for DR intelligent diagnostic technology. Methods Five hundred color fundus photographs of diabetic patients were selected. DR severity varied from grade 0 to 4, with 100 photographs for each grade. Following that, these were diagnosed by both ophthalmologists and the intelligent technology, the results of which were compared by applying the evaluation system. The system includes primary, intermediate, and advanced evaluations, of which the intermediate evaluation incorporated two methods. Main evaluation indicators were sensitivity, specificity, and kappa value. Results The AI technology diagnosed 93 photographs with no DR, 107 with mild non-proliferative DR (NPDR), 107 with moderate NPDR, 108 with severe NPDR, and 85 with proliferative DR (PDR). The sensitivity, specificity, and kappa value of the AI diagnoses in the primary evaluation were 98.8%, 88.0%, and 0.89, respectively. According to method 1 of the intermediate evaluation, the sensitivity of AI diagnosis was 98.0%, specificity 97.0%, and the kappa value 0.95. In method 2 of the intermediate evaluation, the sensitivity of AI diagnosis was 95.5%, the specificity 99.3%, and kappa value 0.95. In the advanced evaluation, the kappa value of the intelligent diagnosis was 0.86. Conclusions This article proposes an evaluation system for color fundus photograph-based intelligent diagnostic technology of DR and demonstrates an application of this system in a clinical setting. The results from this evaluation system serve as the basis for the selection of scenarios in which DR intelligent diagnostic technology can be applied.
引用
收藏
页码:1811 / 1822
页数:12
相关论文
共 50 条
  • [21] A Web-based telemedicine system for diabetic retinopathy screening using digital fundus photography
    Wei, JC
    Valentino, DJ
    Bell, DS
    Baker, RS
    TELEMEDICINE JOURNAL AND E-HEALTH, 2006, 12 (01): : 50 - 57
  • [22] Research on Trustworthiness Evaluation Technology of Aircraft Intelligent Navigation System
    Sun, Jing
    Zhang, Xiujian
    Zhang, Pengcheng
    Zhang, Yongchao
    2020 IEEE 7TH INTERNATIONAL WORKSHOP ON METROLOGY FOR AEROSPACE (METROAEROSPACE), 2020, : 233 - 237
  • [23] CURVELET FUSION ENHACEMENT BASED EVALUATION OF DIABETIC RETINOPATHY BY THE IDENTIFICATION OF EXUDATES IN OPTIC COLOR FUNDUS IMAGES
    Bhargavi, V. Ratna
    Senapati, Ranjan K.
    BIOMEDICAL ENGINEERING-APPLICATIONS BASIS COMMUNICATIONS, 2016, 28 (06):
  • [24] Diagnostic assistance system to detect progression of diabetic retinopathy (DR) in fundus images of follow up examinations of patients with diabetes
    Dietzel, Alexander
    Schanner, Carolin
    Falck, Aura
    Hautala, Nina
    INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE, 2018, 59 (09)
  • [25] Evaluation of a System for Automatic Detection of Diabetic Retinopathy From Color Fundus Photographs in a Large Population of Patients With Diabetes
    Abramoff, Michael D.
    Niemeijer, Meindert
    Suttorp-Schulten, Maria S. A.
    Viergever, Max A.
    Russell, Stephen R.
    van Ginneken, Bram
    DIABETES CARE, 2008, 31 (02) : 193 - 198
  • [26] Diagnostic accuracy of non-mydriatic fundus camera for screening of diabetic retinopathy: A hospital based observational study in Pakistan
    Fahadullah, Muhammad
    Memon, Nasir Ahmed
    Salim, Sumaira
    Ahsan, Shahid
    Fahim, Muhammad Faisal
    Mumtaz, Seema Nigah
    Shaikh, Sikander Ali
    Memon, Muhammad Saleh
    JOURNAL OF THE PAKISTAN MEDICAL ASSOCIATION, 2019, 69 (03) : 378 - 382
  • [27] The diagnostic accuracy of an intelligent and automated fundus disease image assessment system with lesion quantitative function (SmartEye) in diabetic patients
    Yi Xu
    Yongyi Wang
    Bin Liu
    Lin Tang
    Liangqing Lv
    Xin Ke
    Saiguang Ling
    Lina Lu
    Haidong Zou
    BMC Ophthalmology, 19
  • [28] The diagnostic accuracy of an intelligent and automated fundus disease image assessment system with lesion quantitative function (SmartEye) in diabetic patients
    Xu, Yi
    Wang, Yongyi
    Liu, Bin
    Tang, Lin
    Lv, Liangqing
    Ke, Xin
    Ling, Saiguang
    Lu, Lina
    Zou, Haidong
    BMC OPHTHALMOLOGY, 2019, 19 (01)
  • [29] Screening for Diabetic Retinopathy Using an Automated Diagnostic System Based on Deep Learning: Diagnostic Accuracy Assessment
    Rego, Silvia
    Dutra-Medeiros, Marco
    Soares, Filipe
    Monteiro-Soares, Matilde
    OPHTHALMOLOGICA, 2021, 244 (03) : 250 - 257
  • [30] The Intelligent Parking System Based on ZigBee Technology Research and Design
    Hao, Weijun
    Zhao, Yougui
    2016 3RD INTERNATIONAL CONFERENCE ON SMART MATERIALS AND NANOTECHNOLOGY IN ENGINEERING (SMNE 2016), 2016, : 256 - 258