AIROGS: Artificial Intelligence for Robust Glaucoma Screening Challenge

被引：13

作者：

de Vente, Coen ^{[1
,2
,3
]}

Vermeer, Koenraad A. ^{[4
]}

Jaccard, Nicolas ^{[5
]}

Wang, He ^{[6
,7
]}

Sun, Hongyi ^{[8
]}

Khader, Firas ^{[9
]}

Truhn, Daniel ^{[9
]}

Aimyshev, Temirgali ^{[10
]}

Zhanibekuly, Yerkebulan ^{[10
]}

Le, Tien-Dung ^{[11
]}

Galdran, Adrian ^{[12
,13
]}

Ballester, Miguel Angel Gonzalez ^{[12
,14
,24
]}

Carneiro, Gustavo ^{[13
,15
]}

Devika, R. G. ^{[16
]}

Sethumadhavan, Hrishikesh Panikkasseril ^{[17
]}

Puthussery, Densen ^{[17
]}

Liu, Hong ^{[18
]}

Yang, Zekang ^{[18
]}

Kondo, Satoshi ^{[19
]}

Kasai, Satoshi ^{[20
]}

Wang, Edward ^{[21
]}

Durvasula, Ashritha ^{[21
]}

Heras, Jonathan ^{[22
]}

Zapata, Miguel Angel ^{[23
]}

Araujo, Teresa ^{[25
]}

Aresta, Guilherme ^{[25
]}

Bogunovic, Hrvoje ^{[25
]}

Arikan, Mustafa ^{[26
]}

Lee, Yeong Chan ^{[27
]}

Cho, Hyun Bin ^{[28
]}

Choi, Yoon Ho ^{[28
,29
]}

Qayyum, Abdul ^{[30
]}

Razzak, Imran ^{[31
]}

van Ginneken, Bram ^{[3
]}

Lemij, Hans G.

Sanchez, Clara I. ^{[1
,2
]}

机构：

[1] Univ Amsterdam, Informat Inst, Quantitat Healthcare Anal QurAI Grp, NL-1098 XH Amsterdam, Netherlands

[2] Amsterdam UMC Locatie AMC, Dept Biomed Engn & Phys, NL-1105 AZ Amsterdam, Noord Holland, Netherlands

[3] Radboudumc, Dept Radiol & Nucl Med, Diagnost Image Anal Grp DIAG, NL-6500 HB Nijmegen, Gelderland, Netherlands

[4] Rotterdam Eye Hosp, Rotterdam Ophthalm Inst, NL-3011 BH Rotterdam, Netherlands

[5] Project Orbis Int Inc, New York, NY 10017 USA

[6] Peking Union Med Coll Hosp, Beijing 100730, Peoples R China

[7] Capital Med Univ, Xuanwu Hosp, Beijing 100053, Peoples R China

[8] Tsinghua Univ, Dept Automat, Beijing 100190, Peoples R China

[9] Univ Hosp Aachen, Dept Diagnost & Intervent Radiol, D-52074 Aachen, Germany

[10] CMC Technol LLP, Z05T0B8, Nur Sultan, Kazakhstan

[11] KBC, B-1080 Brussels, Belgium

[12] Univ Pompeu Fabra, Dept Tecnol Informacio & Comunicac DTIC, Barcelona 08018, Spain

[13] Univ Adelaide, Australian Inst Machine Learning AIML, Adelaide, SA 5000, Australia

[14] ICREA, Barcelona 08010, Spain

[15] Univ Surrey, Ctr Vis Speech & Signal Proc, Guildford GU2 7XH, England

[16] Coll Engn Trivandrum, Thiruvananthapuram 695016, India

[17] Founding Minds Software, Thiruvananthapuram 682030, India

[18] Chinese Acad Sci, Inst Comp Technol, Beijing 100190, Peoples R China

[19] Muroran Inst Technol, Muroran 0508585, Japan

[20] Niigata Univ Hlth & Welf, Niigata 9503102, Japan

[21] Univ Western Ontario, Schulich Sch Med & Dent, London, ON N6A 5C1, Canada

[22] Univ La Rioja, Dept Math & Comp Sci, Logrono 26004, Spain

[23] Hosp Valle De Hebron, Sant Cugat Del Valles 08195, Spain

[24] UPRetina, Barcelona 08195, Spain

[25] Med Univ Vienna, Dept Ophthalmol & Optometry, Christian Doppler Lab Artificial Intelligence Reti, A-1090 Vienna, Austria

[26] UCL, Inst Ophthalmol, London EC1V 9EL, England

[27] Samsung Med Ctr, Res Inst Future Med, Seoul 06351, South Korea

[28] Sungkyunkwan Univ, Samsung Med Ctr, Samsung Adv Inst Hlth Sci & Technol SAIHST, Dept Digital Hlth, Seoul 06351, South Korea

[29] Mayo Clin, Dept Artificial Intelligence & Informat, Jacksonville, FL 32224 USA

[30] Kings Coll London, Dept Biomed Engn, London, England

[31] Univ New South Wales, Sch Comp Sci & Engn, Sydney, NSW 3125, Australia

来源：

IEEE TRANSACTIONS ON MEDICAL IMAGING | 2024年 / 43卷 / 01期

关键词：

Color fundus photography; glaucoma screening; out-of-distribution detection; retina; robustness; LIFE VISUAL IMPAIRMENT; PREVALENCE;

D O I：

10.1109/TMI.2023.3313786

中图分类号：

TP39 [计算机的应用];

学科分类号：

081203 ; 0835 ;

摘要：

The early detection of glaucoma is essential in preventing visual impairment. Artificial intelligence (AI) can be used to analyze color fundus photographs (CFPs) in a cost-effective manner, making glaucoma screening more accessible. While AI models for glaucoma screening from CFPs have shown promising results in laboratory settings, their performance decreases significantly in real-world scenarios due to the presence of out-of-distribution and low-quality images. To address this issue, we propose the Artificial Intelligence for Robust Glaucoma Screening (AIROGS) challenge. This challenge includes a large dataset of around 113,000 images from about 60,000 patients and 500 different screening centers, and encourages the development of algorithms that are robust to ungradable and unexpected input data. We evaluated solutions from 14 teams in this paper and found that the best teams performed similarly to a set of 20 expert ophthalmologists and optometrists. The highest-scoring team achieved an area under the receiver operating characteristic curve of 0.99 (95% CI: 0.98-0.99) for detecting ungradable images on-the-fly. Additionally, many of the algorithms showed robust performance when tested on three other publicly available datasets. These results demonstrate the feasibility of robust AI-enabled glaucoma screening.

引用

页码：542 / 557

页数：16

共 85 条

[1]

Aimyshev T., 2022, Glaucoma Detection Algorithm for the Artificial Intelligence for Robust Glaucoma Screening Challenge

[2]

[Anonymous], 2017, P AAAI C ART INT

[3]

Araujo T., 2022, P IEEE INT S BIOM IM, P1

[4]

Arikan M., 2022, Multi -Model Ensemble for Robust Glaucoma Screening

[5] A Human-Centered Evaluation of a Deep Learning System Deployed in Clinics for the Detection of Diabetic Retinopathy [J].

Beede, Emma ;

Baylor, Elizabeth ;

Hersch, Fred ;

Iurchenko, Anna ;

Wilcox, Lauren ;

Ruamviboonsuk, Paisan ;

Vardoulakis, Laura M. .

PROCEEDINGS OF THE 2020 CHI CONFERENCE ON HUMAN FACTORS IN COMPUTING SYSTEMS (CHI'20), 2020,

[6]

Bello I, 2021, ADV NEUR IN, V34

[7]

Bibas K, 2021, ADV NEUR IN

[8]

Cao Hu, 2023, Computer Vision - ECCV 2022 Workshops: Proceedings. Lecture Notes in Computer Science (13803), P205, DOI 10.1007/978-3-031-25066-8_9

[9]

Chen XY, 2015, IEEE ENG MED BIO, P5260, DOI 10.1109/EMBC.2015.7319578

[10]

Cuadros Jorge, 2009, J Diabetes Sci Technol, V3, P509

← 1 2 3 4 5 6 7 8 9 →