Auto-classification of Retinal Diseases in the Limit of Sparse Data Using a Two-Streams Machine Learning Model

C. H. Huck Yang; Fangyu Liu; Jia-Hong Huang; Meng Tian; M. D. I-Hung Lin; Yi Chieh Liu; Hiromasa Morikawa; Hao Hsiang Yang; Jesper Tegner

doi:10.1007/978-3-030-21074-8_28

Auto-classification of Retinal Diseases in the Limit of Sparse Data Using a Two-Streams Machine Learning Model

C. H. Huck Yang, Fangyu Liu, Jia-Hong Huang, Meng Tian, M. D. I-Hung Lin, Yi Chieh Liu, Hiromasa Morikawa, Hao Hsiang Yang, Jesper Tegner

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

12 Scopus citations

Abstract

Automatic clinical diagnosis of retinal diseases has emerged as a promising approach to facilitate discovery in areas with limited access to specialists. Based on the fact that fundus structure and vascular disorders are the main characteristics of retinal diseases, we propose a novel visual-assisted diagnosis hybrid model mixing the support vector machine (SVM) and deep neural networks (DNNs). Furthermore, we present a new clinical retina labels collection sorted by the professional ophthalmologist from the educational project Retina Image Bank, called EyeNet, for ophthalmology incorporating 52 retina diseases classes. Using EyeNet, our model achieves 90.40% diagnosis accuracy, and the model performance is comparable to the professional ophthalmologists (https://github.com/huckiyang/EyeNet2).

Original language	English (US)
Title of host publication	Computer Vision – ACCV 2018 Workshops
Publisher	Springer Nature
Pages	323-338
Number of pages	16
ISBN (Print)	9783030210731
DOIs	https://doi.org/10.1007/978-3-030-21074-8_28
State	Published - Jun 19 2019

Access to Document

10.1007/978-3-030-21074-8_28

https://repository.kaust.edu.sa/bitstream/10754/656187/1/1808.05754.pdf

Cite this

@inproceedings{822633ad65e7405bb3bef2d306a50a93,

title = "Auto-classification of Retinal Diseases in the Limit of Sparse Data Using a Two-Streams Machine Learning Model",

abstract = "Automatic clinical diagnosis of retinal diseases has emerged as a promising approach to facilitate discovery in areas with limited access to specialists. Based on the fact that fundus structure and vascular disorders are the main characteristics of retinal diseases, we propose a novel visual-assisted diagnosis hybrid model mixing the support vector machine (SVM) and deep neural networks (DNNs). Furthermore, we present a new clinical retina labels collection sorted by the professional ophthalmologist from the educational project Retina Image Bank, called EyeNet, for ophthalmology incorporating 52 retina diseases classes. Using EyeNet, our model achieves 90.40% diagnosis accuracy, and the model performance is comparable to the professional ophthalmologists (https://github.com/huckiyang/EyeNet2).",

author = "{Huck Yang}, {C. H.} and Fangyu Liu and Jia-Hong Huang and Meng Tian and {I-Hung Lin}, {M. D.} and Liu, {Yi Chieh} and Hiromasa Morikawa and Yang, {Hao Hsiang} and Jesper Tegner",

note = "KAUST Repository Item: Exported on 2020-10-01",

year = "2019",

month = jun,

day = "19",

doi = "10.1007/978-3-030-21074-8_28",

language = "English (US)",

isbn = "9783030210731",

pages = "323--338",

booktitle = "Computer Vision – ACCV 2018 Workshops",

publisher = "Springer Nature",

}

TY - GEN

T1 - Auto-classification of Retinal Diseases in the Limit of Sparse Data Using a Two-Streams Machine Learning Model

AU - Huck Yang, C. H.

AU - Liu, Fangyu

AU - Huang, Jia-Hong

AU - Tian, Meng

AU - I-Hung Lin, M. D.

AU - Liu, Yi Chieh

AU - Morikawa, Hiromasa

AU - Yang, Hao Hsiang

AU - Tegner, Jesper

N1 - KAUST Repository Item: Exported on 2020-10-01

PY - 2019/6/19

Y1 - 2019/6/19

N2 - Automatic clinical diagnosis of retinal diseases has emerged as a promising approach to facilitate discovery in areas with limited access to specialists. Based on the fact that fundus structure and vascular disorders are the main characteristics of retinal diseases, we propose a novel visual-assisted diagnosis hybrid model mixing the support vector machine (SVM) and deep neural networks (DNNs). Furthermore, we present a new clinical retina labels collection sorted by the professional ophthalmologist from the educational project Retina Image Bank, called EyeNet, for ophthalmology incorporating 52 retina diseases classes. Using EyeNet, our model achieves 90.40% diagnosis accuracy, and the model performance is comparable to the professional ophthalmologists (https://github.com/huckiyang/EyeNet2).

AB - Automatic clinical diagnosis of retinal diseases has emerged as a promising approach to facilitate discovery in areas with limited access to specialists. Based on the fact that fundus structure and vascular disorders are the main characteristics of retinal diseases, we propose a novel visual-assisted diagnosis hybrid model mixing the support vector machine (SVM) and deep neural networks (DNNs). Furthermore, we present a new clinical retina labels collection sorted by the professional ophthalmologist from the educational project Retina Image Bank, called EyeNet, for ophthalmology incorporating 52 retina diseases classes. Using EyeNet, our model achieves 90.40% diagnosis accuracy, and the model performance is comparable to the professional ophthalmologists (https://github.com/huckiyang/EyeNet2).

UR - http://hdl.handle.net/10754/656187

UR - http://link.springer.com/10.1007/978-3-030-21074-8_28

UR - http://www.scopus.com/inward/record.url?scp=85068441690&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-21074-8_28

DO - 10.1007/978-3-030-21074-8_28

M3 - Conference contribution

SN - 9783030210731

SP - 323

EP - 338

BT - Computer Vision – ACCV 2018 Workshops

PB - Springer Nature

ER -

Auto-classification of Retinal Diseases in the Limit of Sparse Data Using a Two-Streams Machine Learning Model

Abstract

Access to Document

Other files and links

Fingerprint

Cite this