• Development of a cloud-based e...

Treffer: Development of a cloud-based estimator for analysing the disc-fovea angle in fundus images using a stacking ensemble model.

Title:
Development of a cloud-based estimator for analysing the disc-fovea angle in fundus images using a stacking ensemble model.
Authors:
Hirota M; Department of Orthoptics, Teikyo University, Itabashi, Japan hirota.ortho@med.teikyo-u.ac.jp.; Department of Ophthalmology, Teikyo University, Itabashi, Japan.; Graduate Degree Program of Comprehensive Applied Data Science, Teikyo University, Itabashi, Japan.; Graduate Degree Program of Health Data Science, Teikyo University, Itabashi, Japan., Watanabe M; Division of Orthoptics, Graduate School of Medical Care and Technology Teikyo University, Itabashi, Japan., Sasaki K; Department of Orthoptics, Teikyo University, Itabashi, Japan.; Department of Ophthalmology, Teikyo University, Itabashi, Japan., Kato K; Department of Orthoptics, Teikyo University, Itabashi, Japan., Usui C; Department of Orthoptics, Teikyo University, Itabashi, Japan.; Department of Ophthalmology, Teikyo University, Itabashi, Japan., Mizuno Y; Department of Ophthalmology, Teikyo University, Itabashi, Japan., Hayashi T; Department of Orthoptics, Teikyo University, Itabashi, Japan.; Department of Ophthalmology, Teikyo University, Itabashi, Japan., Inoue Y; Department of Ophthalmology, Teikyo University, Itabashi, Japan., Mizota A; Department of Ophthalmology, Teikyo University, Itabashi, Japan.; Ophthalmology, Nishikasai Inouye Eye Hospital, Edogawa, Japan.
Source:
BMJ open ophthalmology [BMJ Open Ophthalmol] 2026 Jan 14; Vol. 11 (1). Date of Electronic Publication: 2026 Jan 14.
Publication Type:
Journal Article
Language:
English
Journal Info:
Publisher: BMJ Publishing Group Ltd Country of Publication: England NLM ID: 101714806 Publication Model: Electronic Cited Medium: Internet ISSN: 2397-3269 (Electronic) Linking ISSN: 23973269 NLM ISO Abbreviation: BMJ Open Ophthalmol Subsets: MEDLINE
Imprint Name(s):
Original Publication: [London] : BMJ Publishing Group Ltd., [2017]-
MeSH Terms:
Optic Disk*/pathology , Optic Disk*/diagnostic imaging , Cloud Computing* , Image Processing, Computer-Assisted*/methods, Humans ; Retrospective Studies ; Fundus Oculi ; Female ; Male ; Adult ; Algorithms ; Middle Aged ; Young Adult ; Neural Networks, Computer ; Machine Learning ; Software
References:
Invest Ophthalmol Vis Sci. 2014 Nov 25;56(1):156-63. (PMID: 25425302)
J Digit Imaging. 2018 Dec;31(6):923-928. (PMID: 29948436)
Int Ophthalmol. 2020 Nov;40(11):3013-3022. (PMID: 32594350)
Eur J Ophthalmol. 2021 Mar;31(2):704-708. (PMID: 32054328)
PLoS One. 2020 Dec 22;15(12):e0244230. (PMID: 33351818)
BMJ Open Ophthalmol. 2020 Dec 24;5(1):e000595. (PMID: 33409371)
Transl Vis Sci Technol. 2021 May 3;10(6):1. (PMID: 34111248)
Cochrane Database Syst Rev. 2017 Nov 27;11:CD012447. (PMID: 29178265)
Transl Vis Sci Technol. 2020 Feb 25;9(3):27. (PMID: 32742757)
IEEE Trans Pattern Anal Mach Intell. 2017 Jun;39(6):1137-1149. (PMID: 27295650)
J Pers Med. 2021 Apr 21;11(5):. (PMID: 33918998)
J Pediatr Ophthalmol Strabismus. 1984 Jul-Aug;21(4):126-32. (PMID: 6470908)
J AAPOS. 2017 Aug;21(4):305-308. (PMID: 28666773)
Stat Methods Med Res. 1999 Jun;8(2):135-60. (PMID: 10501650)
Acta Ophthalmol. 2018 Nov;96(7):e840-e845. (PMID: 29369505)
J AAPOS. 2018 Apr;22(2):146-148.e1. (PMID: 29317291)
Ophthalmology. 1982 Jan;89(1):58-62. (PMID: 7070775)
J Glaucoma. 2020 Oct;29(10):964-969. (PMID: 32618803)
Am Orthopt J. 2014;64:76-80. (PMID: 25313115)
JAMA Ophthalmol. 2013 May;131(5):619-25. (PMID: 23471194)
Eye (Lond). 2005 Aug;19(8):873-8. (PMID: 15731772)
Contributed Indexing:
Keywords: Imaging; Strabismus
Entry Date(s):
Date Created: 20260114 Date Completed: 20260114 Latest Revision: 20260121
Update Code:
20260121
PubMed Central ID:
PMC12815236
DOI:
10.1136/bmjophth-2025-002225
PMID:
41535013
Database:
MEDLINE

Weitere Informationen

Purpose: To develop a cloud-based software for analysing the disc-fovea angle (DFA) in fundus images using the stacking ensemble model and to evaluate the accuracy of the objective DFA assessment between automatic and manual analyses.
Methods: A total of 682 fundus images (573 and 109 images of healthy individuals and patients with cyclotropia, respectively) were included in this retrospective study. For manual DFA analysis, two examiners analysed the DFA using web-based analysis software for the DFA, which was defined as the average value between the two examiners. For automatic DFA analysis, the stacking ensemble model on the cloud web integrated a single-shot multi-box detector to identify the fovea and centre of the optic nerve head, along with a single convolutional neural network combined with five machine learning algorithms.
Results: The DFA did not significantly vary between the automatic (7.25° ± 4.93°) and manual analyses (7.27° ± 5.07°) (p=0.52). The time required to analyse the DFA in a single fundus image was significantly shorter with the automatic analysis (0.430±0.240 s) than with the manual analysis (15.794±1.558 s) (p<0.001).
Conclusion: These findings suggest that automatic analysis using the stacking ensemble model is useful for determining the DFA in clinical settings.
(© Author(s) (or their employer(s)) 2026. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ Group.)

Competing interests: None declared.