Classification of Bacterial Keratitis Activity with Patch-Based Deep Learning Using Three Anterior Segment Images

Sung Ho Jung; Yeokyoung Won; Won Seok Song; Ju Hwan Lee; Hakje Yoo; Dong Hui Lim

doi:10.1109/ISBI56570.2024.10635598

Classification of Bacterial Keratitis Activity with Patch-Based Deep Learning Using Three Anterior Segment Images

Sung Ho Jung, Yeokyoung Won, Won Seok Song, Ju Hwan Lee, Hakje Yoo, Dong Hui Lim

Department of Ophthalmology

Sungkyunkwan University

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

Abstract

Bacterial keratitis is one of the common corneal diseases. Without timely and appropriate treatment, it can lead to complications such as vision reduction and perforation may occur, and in severe cases, it may even lead to blindness. In this study, aim to develop an artificial intelligence model for activity classification in bacterial keratitis using three types of anterior segment images: broad, slit, and scatter. By applying the patch technique, the highest AUROC of the model trained from the original was improved from 0.802 to 0.897. Performance experiments based on image combinations demonstrated that the model using only slit images showed the best results.

Original language	English
Title of host publication	IEEE International Symposium on Biomedical Imaging, ISBI 2024 - Conference Proceedings
Publisher	IEEE Computer Society
ISBN (Electronic)	9798350313338
DOIs	https://doi.org/10.1109/ISBI56570.2024.10635598
State	Published - 2024
Event	21st IEEE International Symposium on Biomedical Imaging, ISBI 2024 - Athens, Greece Duration: 27 May 2024 → 30 May 2024

Publication series

Name	Proceedings - International Symposium on Biomedical Imaging
ISSN (Print)	1945-7928
ISSN (Electronic)	1945-8452

Conference

Conference	21st IEEE International Symposium on Biomedical Imaging, ISBI 2024
Country/Territory	Greece
City	Athens
Period	27/05/24 → 30/05/24

Keywords

anterior segment image
Bacterial keratitis
deep learning
image classification

Access to Document

10.1109/ISBI56570.2024.10635598

Cite this

Jung, S. H., Won, Y., Song, W. S., Hwan Lee, J., Yoo, H., & Lim, D. H. (2024). Classification of Bacterial Keratitis Activity with Patch-Based Deep Learning Using Three Anterior Segment Images. In IEEE International Symposium on Biomedical Imaging, ISBI 2024 - Conference Proceedings (Proceedings - International Symposium on Biomedical Imaging). IEEE Computer Society. https://doi.org/10.1109/ISBI56570.2024.10635598

@inproceedings{860a12c8c165409bb748f19d4e5bad93,

title = "Classification of Bacterial Keratitis Activity with Patch-Based Deep Learning Using Three Anterior Segment Images",

abstract = "Bacterial keratitis is one of the common corneal diseases. Without timely and appropriate treatment, it can lead to complications such as vision reduction and perforation may occur, and in severe cases, it may even lead to blindness. In this study, aim to develop an artificial intelligence model for activity classification in bacterial keratitis using three types of anterior segment images: broad, slit, and scatter. By applying the patch technique, the highest AUROC of the model trained from the original was improved from 0.802 to 0.897. Performance experiments based on image combinations demonstrated that the model using only slit images showed the best results.",

keywords = "anterior segment image, Bacterial keratitis, deep learning, image classification",

author = "Jung, \{Sung Ho\} and Yeokyoung Won and Song, \{Won Seok\} and \{Hwan Lee\}, Ju and Hakje Yoo and Lim, \{Dong Hui\}",

note = "Publisher Copyright: {\textcopyright} 2024 IEEE.; 21st IEEE International Symposium on Biomedical Imaging, ISBI 2024 ; Conference date: 27-05-2024 Through 30-05-2024",

year = "2024",

doi = "10.1109/ISBI56570.2024.10635598",

language = "English",

series = "Proceedings - International Symposium on Biomedical Imaging",

publisher = "IEEE Computer Society",

booktitle = "IEEE International Symposium on Biomedical Imaging, ISBI 2024 - Conference Proceedings",

}

Jung, SH, Won, Y, Song, WS, Hwan Lee, J, Yoo, H & Lim, DH 2024, Classification of Bacterial Keratitis Activity with Patch-Based Deep Learning Using Three Anterior Segment Images. in IEEE International Symposium on Biomedical Imaging, ISBI 2024 - Conference Proceedings. Proceedings - International Symposium on Biomedical Imaging, IEEE Computer Society, 21st IEEE International Symposium on Biomedical Imaging, ISBI 2024, Athens, Greece, 27/05/24. https://doi.org/10.1109/ISBI56570.2024.10635598

Classification of Bacterial Keratitis Activity with Patch-Based Deep Learning Using Three Anterior Segment Images. / Jung, Sung Ho; Won, Yeokyoung; Song, Won Seok et al.
IEEE International Symposium on Biomedical Imaging, ISBI 2024 - Conference Proceedings. IEEE Computer Society, 2024. (Proceedings - International Symposium on Biomedical Imaging).

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

TY - GEN

T1 - Classification of Bacterial Keratitis Activity with Patch-Based Deep Learning Using Three Anterior Segment Images

AU - Jung, Sung Ho

AU - Won, Yeokyoung

AU - Song, Won Seok

AU - Hwan Lee, Ju

AU - Yoo, Hakje

AU - Lim, Dong Hui

PY - 2024

Y1 - 2024

N2 - Bacterial keratitis is one of the common corneal diseases. Without timely and appropriate treatment, it can lead to complications such as vision reduction and perforation may occur, and in severe cases, it may even lead to blindness. In this study, aim to develop an artificial intelligence model for activity classification in bacterial keratitis using three types of anterior segment images: broad, slit, and scatter. By applying the patch technique, the highest AUROC of the model trained from the original was improved from 0.802 to 0.897. Performance experiments based on image combinations demonstrated that the model using only slit images showed the best results.

AB - Bacterial keratitis is one of the common corneal diseases. Without timely and appropriate treatment, it can lead to complications such as vision reduction and perforation may occur, and in severe cases, it may even lead to blindness. In this study, aim to develop an artificial intelligence model for activity classification in bacterial keratitis using three types of anterior segment images: broad, slit, and scatter. By applying the patch technique, the highest AUROC of the model trained from the original was improved from 0.802 to 0.897. Performance experiments based on image combinations demonstrated that the model using only slit images showed the best results.

KW - anterior segment image

KW - Bacterial keratitis

KW - deep learning

KW - image classification

UR - https://www.scopus.com/pages/publications/85203371550

U2 - 10.1109/ISBI56570.2024.10635598

DO - 10.1109/ISBI56570.2024.10635598

M3 - Conference contribution

AN - SCOPUS:85203371550

T3 - Proceedings - International Symposium on Biomedical Imaging

BT - IEEE International Symposium on Biomedical Imaging, ISBI 2024 - Conference Proceedings

PB - IEEE Computer Society

T2 - 21st IEEE International Symposium on Biomedical Imaging, ISBI 2024

Y2 - 27 May 2024 through 30 May 2024

ER -

Jung SH, Won Y, Song WS, Hwan Lee J, Yoo H, Lim DH. Classification of Bacterial Keratitis Activity with Patch-Based Deep Learning Using Three Anterior Segment Images. In IEEE International Symposium on Biomedical Imaging, ISBI 2024 - Conference Proceedings. IEEE Computer Society. 2024. (Proceedings - International Symposium on Biomedical Imaging). doi: 10.1109/ISBI56570.2024.10635598

Classification of Bacterial Keratitis Activity with Patch-Based Deep Learning Using Three Anterior Segment Images

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this