Towards AI based Ophthalmological Screening through Ultra-widefield Fundus Image to Conventional Fundus Image Translation

Pham Van Nguyen; Duc Tai Le; Su Jeong Song; Hyunseung Choo

doi:10.1109/IMCOM53663.2022.9721717

Towards AI based Ophthalmological Screening through Ultra-widefield Fundus Image to Conventional Fundus Image Translation

Pham Van Nguyen
, Duc Tai Le
, Su Jeong Song
, Hyunseung Choo

Sungkyunkwan University

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

1 Scopus citations

Abstract

Conventional fundus image (CFI) has been the most popular modality used in ophthalmological diagnosis. However, taking the CFI is costly and a burden for patients since it requires pupil to dilate. Recent research have shifted more attention towards ultra-widefield fundus image (UFI) which includes a larger area, is cheaper and easier to take. Although features of an CFI can be found in a corresponding UFI, the use of UFIs for eye diagnosis is still limited due to the low contrast and inconsistent background color. The recent advancements in deep learning promote UFI-to-CFI translation to be a promising direction for an early ophthalmological screening. Existing methods cannot deal with low-quality image samples and their outputs usually have low brightness. In this paper, we outperform other works by a novel framework which tackles above problems. In this framework, we deploy an object detector and an illumination estimator to refine input samples of an attention-aided cydeGAN model which is used to generate the CFI. Numerous experiments state that 98.8% of the generated CFIs are recognized as good quality which shows the suitability of our framework for an ophthalmological screening system.

Original language	English
Title of host publication	Proceedings of the 2022 16th International Conference on Ubiquitous Information Management and Communication, IMCOM 2022
Editors	Sukhan Lee, Hyunseung Choo, Roslan Ismail
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781665426787
DOIs	https://doi.org/10.1109/IMCOM53663.2022.9721717
State	Published - 2022
Event	16th International Conference on Ubiquitous Information Management and Communication, IMCOM 2022 - Seoul, Korea, Republic of Duration: 3 Jan 2022 → 5 Jan 2022

Publication series

Name	Proceedings of the 2022 16th International Conference on Ubiquitous Information Management and Communication, IMCOM 2022

Conference

Conference	16th International Conference on Ubiquitous Information Management and Communication, IMCOM 2022
Country/Territory	Korea, Republic of
City	Seoul
Period	3/01/22 → 5/01/22

Keywords

Conventional fundus image
deep learning
image-to-image translation
Ultra wide-field fundus image

Access to Document

10.1109/IMCOM53663.2022.9721717

Cite this

Van Nguyen, P., Le, D. T., Song, S. J., & Choo, H. (2022). Towards AI based Ophthalmological Screening through Ultra-widefield Fundus Image to Conventional Fundus Image Translation. In S. Lee, H. Choo, & R. Ismail (Eds.), Proceedings of the 2022 16th International Conference on Ubiquitous Information Management and Communication, IMCOM 2022 (Proceedings of the 2022 16th International Conference on Ubiquitous Information Management and Communication, IMCOM 2022). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IMCOM53663.2022.9721717

Van Nguyen, Pham ; Le, Duc Tai ; Song, Su Jeong et al. / Towards AI based Ophthalmological Screening through Ultra-widefield Fundus Image to Conventional Fundus Image Translation. Proceedings of the 2022 16th International Conference on Ubiquitous Information Management and Communication, IMCOM 2022. editor / Sukhan Lee ; Hyunseung Choo ; Roslan Ismail. Institute of Electrical and Electronics Engineers Inc., 2022. (Proceedings of the 2022 16th International Conference on Ubiquitous Information Management and Communication, IMCOM 2022).

@inproceedings{fa0726451aed4b4d901e36616b254b0a,

title = "Towards AI based Ophthalmological Screening through Ultra-widefield Fundus Image to Conventional Fundus Image Translation",

abstract = "Conventional fundus image (CFI) has been the most popular modality used in ophthalmological diagnosis. However, taking the CFI is costly and a burden for patients since it requires pupil to dilate. Recent research have shifted more attention towards ultra-widefield fundus image (UFI) which includes a larger area, is cheaper and easier to take. Although features of an CFI can be found in a corresponding UFI, the use of UFIs for eye diagnosis is still limited due to the low contrast and inconsistent background color. The recent advancements in deep learning promote UFI-to-CFI translation to be a promising direction for an early ophthalmological screening. Existing methods cannot deal with low-quality image samples and their outputs usually have low brightness. In this paper, we outperform other works by a novel framework which tackles above problems. In this framework, we deploy an object detector and an illumination estimator to refine input samples of an attention-aided cydeGAN model which is used to generate the CFI. Numerous experiments state that 98.8\% of the generated CFIs are recognized as good quality which shows the suitability of our framework for an ophthalmological screening system.",

keywords = "Conventional fundus image, deep learning, image-to-image translation, Ultra wide-field fundus image",

author = "\{Van Nguyen\}, Pham and Le, \{Duc Tai\} and Song, \{Su Jeong\} and Hyunseung Choo",

note = "Publisher Copyright: {\textcopyright} 2022 IEEE.; 16th International Conference on Ubiquitous Information Management and Communication, IMCOM 2022 ; Conference date: 03-01-2022 Through 05-01-2022",

year = "2022",

doi = "10.1109/IMCOM53663.2022.9721717",

language = "English",

series = "Proceedings of the 2022 16th International Conference on Ubiquitous Information Management and Communication, IMCOM 2022",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

editor = "Sukhan Lee and Hyunseung Choo and Roslan Ismail",

booktitle = "Proceedings of the 2022 16th International Conference on Ubiquitous Information Management and Communication, IMCOM 2022",

}

Van Nguyen, P, Le, DT, Song, SJ & Choo, H 2022, Towards AI based Ophthalmological Screening through Ultra-widefield Fundus Image to Conventional Fundus Image Translation. in S Lee, H Choo & R Ismail (eds), Proceedings of the 2022 16th International Conference on Ubiquitous Information Management and Communication, IMCOM 2022. Proceedings of the 2022 16th International Conference on Ubiquitous Information Management and Communication, IMCOM 2022, Institute of Electrical and Electronics Engineers Inc., 16th International Conference on Ubiquitous Information Management and Communication, IMCOM 2022, Seoul, Korea, Republic of, 3/01/22. https://doi.org/10.1109/IMCOM53663.2022.9721717

Towards AI based Ophthalmological Screening through Ultra-widefield Fundus Image to Conventional Fundus Image Translation. / Van Nguyen, Pham; Le, Duc Tai; Song, Su Jeong et al.
Proceedings of the 2022 16th International Conference on Ubiquitous Information Management and Communication, IMCOM 2022. ed. / Sukhan Lee; Hyunseung Choo; Roslan Ismail. Institute of Electrical and Electronics Engineers Inc., 2022. (Proceedings of the 2022 16th International Conference on Ubiquitous Information Management and Communication, IMCOM 2022).

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

TY - GEN

T1 - Towards AI based Ophthalmological Screening through Ultra-widefield Fundus Image to Conventional Fundus Image Translation

AU - Van Nguyen, Pham

AU - Le, Duc Tai

AU - Song, Su Jeong

AU - Choo, Hyunseung

PY - 2022

Y1 - 2022

N2 - Conventional fundus image (CFI) has been the most popular modality used in ophthalmological diagnosis. However, taking the CFI is costly and a burden for patients since it requires pupil to dilate. Recent research have shifted more attention towards ultra-widefield fundus image (UFI) which includes a larger area, is cheaper and easier to take. Although features of an CFI can be found in a corresponding UFI, the use of UFIs for eye diagnosis is still limited due to the low contrast and inconsistent background color. The recent advancements in deep learning promote UFI-to-CFI translation to be a promising direction for an early ophthalmological screening. Existing methods cannot deal with low-quality image samples and their outputs usually have low brightness. In this paper, we outperform other works by a novel framework which tackles above problems. In this framework, we deploy an object detector and an illumination estimator to refine input samples of an attention-aided cydeGAN model which is used to generate the CFI. Numerous experiments state that 98.8% of the generated CFIs are recognized as good quality which shows the suitability of our framework for an ophthalmological screening system.

AB - Conventional fundus image (CFI) has been the most popular modality used in ophthalmological diagnosis. However, taking the CFI is costly and a burden for patients since it requires pupil to dilate. Recent research have shifted more attention towards ultra-widefield fundus image (UFI) which includes a larger area, is cheaper and easier to take. Although features of an CFI can be found in a corresponding UFI, the use of UFIs for eye diagnosis is still limited due to the low contrast and inconsistent background color. The recent advancements in deep learning promote UFI-to-CFI translation to be a promising direction for an early ophthalmological screening. Existing methods cannot deal with low-quality image samples and their outputs usually have low brightness. In this paper, we outperform other works by a novel framework which tackles above problems. In this framework, we deploy an object detector and an illumination estimator to refine input samples of an attention-aided cydeGAN model which is used to generate the CFI. Numerous experiments state that 98.8% of the generated CFIs are recognized as good quality which shows the suitability of our framework for an ophthalmological screening system.

KW - Conventional fundus image

KW - deep learning

KW - image-to-image translation

KW - Ultra wide-field fundus image

UR - https://www.scopus.com/pages/publications/85127498006

U2 - 10.1109/IMCOM53663.2022.9721717

DO - 10.1109/IMCOM53663.2022.9721717

M3 - Conference contribution

AN - SCOPUS:85127498006

T3 - Proceedings of the 2022 16th International Conference on Ubiquitous Information Management and Communication, IMCOM 2022

BT - Proceedings of the 2022 16th International Conference on Ubiquitous Information Management and Communication, IMCOM 2022

A2 - Lee, Sukhan

A2 - Choo, Hyunseung

A2 - Ismail, Roslan

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 16th International Conference on Ubiquitous Information Management and Communication, IMCOM 2022

Y2 - 3 January 2022 through 5 January 2022

ER -

Van Nguyen P, Le DT, Song SJ , Choo H. Towards AI based Ophthalmological Screening through Ultra-widefield Fundus Image to Conventional Fundus Image Translation. In Lee S, Choo H, Ismail R, editors, Proceedings of the 2022 16th International Conference on Ubiquitous Information Management and Communication, IMCOM 2022. Institute of Electrical and Electronics Engineers Inc. 2022. (Proceedings of the 2022 16th International Conference on Ubiquitous Information Management and Communication, IMCOM 2022). doi: 10.1109/IMCOM53663.2022.9721717

Towards AI based Ophthalmological Screening through Ultra-widefield Fundus Image to Conventional Fundus Image Translation

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this