Segmentation of macular neovascularization and leakage in fluorescein angiography images in neovascular age-related macular degeneration using deep learning.
Journal
Eye (London, England)
ISSN: 1476-5454
Titre abrégé: Eye (Lond)
Pays: England
ID NLM: 8703986
Informations de publication
Date de publication:
05 2023
05 2023
Historique:
received:
17
02
2022
accepted:
16
06
2022
revised:
31
05
2022
medline:
11
5
2023
pubmed:
2
7
2022
entrez:
1
7
2022
Statut:
ppublish
Résumé
We aim to develop an objective fully automated Artificial intelligence (AI) algorithm for MNV lesion size and leakage area segmentation on fluorescein angiography (FA) in patients with neovascular age-related macular degeneration (nAMD). Two FA image datasets collected form large prospective multicentre trials consisting of 4710 images from 513 patients and 4558 images from 514 patients were used to develop and evaluate a deep learning-based algorithm to detect CNV lesion size and leakage area automatically. Manual segmentation of was performed by certified FA graders of the Vienna Reading Center. Precision, Recall and F1 score between AI predictions and manual annotations were computed. In addition, two masked retina experts conducted a clinical-applicability evaluation, comparing the quality of AI based and manual segmentations. For CNV lesion size and leakage area segmentation, we obtained F1 scores of 0.73 and 0.65, respectively. Expert review resulted in a slight preference for the automated segmentations in both datasets. The quality of automated segmentations was slightly more often judged as good compared to manual annotations. CNV lesion size and leakage area can be segmented by our automated model at human-level performance, its output being well-accepted during clinical applicability testing. The results provide proof-of-concept that an automated deep learning approach can improve efficacy of objective biomarker analysis in FA images and will be well-suited for clinical application.
Sections du résumé
BACKGROUND/OBJECTIVES
We aim to develop an objective fully automated Artificial intelligence (AI) algorithm for MNV lesion size and leakage area segmentation on fluorescein angiography (FA) in patients with neovascular age-related macular degeneration (nAMD).
SUBJECTS/METHODS
Two FA image datasets collected form large prospective multicentre trials consisting of 4710 images from 513 patients and 4558 images from 514 patients were used to develop and evaluate a deep learning-based algorithm to detect CNV lesion size and leakage area automatically. Manual segmentation of was performed by certified FA graders of the Vienna Reading Center. Precision, Recall and F1 score between AI predictions and manual annotations were computed. In addition, two masked retina experts conducted a clinical-applicability evaluation, comparing the quality of AI based and manual segmentations.
RESULTS
For CNV lesion size and leakage area segmentation, we obtained F1 scores of 0.73 and 0.65, respectively. Expert review resulted in a slight preference for the automated segmentations in both datasets. The quality of automated segmentations was slightly more often judged as good compared to manual annotations.
CONCLUSIONS
CNV lesion size and leakage area can be segmented by our automated model at human-level performance, its output being well-accepted during clinical applicability testing. The results provide proof-of-concept that an automated deep learning approach can improve efficacy of objective biomarker analysis in FA images and will be well-suited for clinical application.
Identifiants
pubmed: 35778604
doi: 10.1038/s41433-022-02156-6
pii: 10.1038/s41433-022-02156-6
pmc: PMC10169785
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
1439-1444Informations de copyright
© 2022. The Author(s).
Références
Ophthalmology. 2008 Jun;115(6):1026-31
pubmed: 17981333
J Am Coll Radiol. 2014 Oct;11(10):974-8
pubmed: 24889471
Retina. 2002 Dec;22(6):683-90
pubmed: 12476092
IEEE Trans Image Process. 2020 May 08;:
pubmed: 32396087
Transl Vis Sci Technol. 2020 Sep 17;9(2):52
pubmed: 32995069
Prog Retin Eye Res. 2018 Nov;67:1-29
pubmed: 30076935
Br J Ophthalmol. 2014 Sep;98(9):1144-67
pubmed: 25136079
Biomed Opt Express. 2020 Jan 14;11(2):927-944
pubmed: 32133230
Neuroimage. 2021 Jun;233:117934
pubmed: 33737246
Invest Ophthalmol Vis Sci. 2006 Dec;47(12):5460-8
pubmed: 17122137
Lancet Glob Health. 2014 Feb;2(2):e106-16
pubmed: 25104651
Med Image Anal. 2020 Jan;59:101557
pubmed: 31677438
Ophthalmic Surg Lasers Imaging Retina. 2015 Jun;46(6):611-6
pubmed: 26172062
Insights Imaging. 2018 Aug;9(4):611-629
pubmed: 29934920
IEEE Trans Biomed Eng. 2013 May;60(5):1439-45
pubmed: 23314765