2020
DOI: 10.1371/journal.pone.0227240
|View full text |Cite
|
Sign up to set email alerts
|

Accuracy of a deep convolutional neural network in the detection of myopic macular diseases using swept-source optical coherence tomography

Abstract: This study examined and compared outcomes of deep learning (DL) in identifying sweptsource optical coherence tomography (OCT) images without myopic macular lesions [i.e., no high myopia (nHM) vs. high myopia (HM)], and OCT images with myopic macular lesions [e.g., myopic choroidal neovascularization (mCNV) and retinoschisis (RS)]. A total of 910 SS-OCT images were included in the study as follows and analyzed by k-fold cross-validation (k = 5) using DL's renowned model, Visual Geometry Group-16: nHM, 146 image… Show more

Help me understand this report
View preprint versions

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
33
0
1

Year Published

2021
2021
2024
2024

Publication Types

Select...
10

Relationship

0
10

Authors

Journals

citations
Cited by 43 publications
(34 citation statements)
references
References 48 publications
0
33
0
1
Order By: Relevance
“… 31 Therefore, it may be more feasible to use other imaging information such as OCT images as training sources for such classifications. 36 …”
Section: Discussionmentioning
confidence: 99%
“… 31 Therefore, it may be more feasible to use other imaging information such as OCT images as training sources for such classifications. 36 …”
Section: Discussionmentioning
confidence: 99%
“…Researchers have developed deep learning models to predict the uncorrected refractive error from OCT images, indicating that OCT could also be used to estimate refractive error [ 130 ]. OCT image-based deep learning algorithms showed robust performances in the detection of myopic maculopathy [ 131 , 132 , 133 ], glaucomatous optic neuropathy [ 134 ], as well as analysis of choroidal features [ 135 , 136 ]. In addition, machine learning of the preoperative AS-OCT metrics is able to provide high predictability of the ICL vault, indicating that it may be helpful for selecting the proper ICL size in clinical practice [ 137 ].…”
Section: Artificial Intelligence In Oct/octa In Myopiamentioning
confidence: 99%
“…Other similar studies have also been reported, such as those identifying the different types of lesions of myopic maculopathy automatically from fundus photographs with DL models ( 40 , 41 ). In addition, OCT macular images were used for the development of CNN models to identify vision-threatening conditions, such as retinoschisis, macular holes and retinal detachment, in adults with high myopia, and the models obtained good sensitivity and AUC scores ( 42 , 43 ).…”
Section: Ai In the Detection And Diagnosis Of Myopiamentioning
confidence: 99%