Automated diagnosis and segmentation of choroidal neovascularization in OCT angiography using deep learning

Wang, Jie; Hormel, Tristan T.; Gao, Linlin; Zang, Pengxiao; Guo, Yukun; Wang, Xiaogang; Bailey, Steven T.; Jia, Yali

doi:10.1364/boe.379977

Cited by 59 publications

(39 citation statements)

References 49 publications

(49 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In vivo histological analysis using OCT scans is probably the most attractive method for studying the disease through artificial intelligence. In fact, these scans are rich in details and can be easily accessed by radiomics [37,38]. Moreover, structural OCT and OCT-A are amongst the most effective non-invasive imaging tools for several retinal diseases.…”

Section: Discussionmentioning

confidence: 99%

Biomarkers in Early Response to Brolucizumab on Pigment Epithelium Detachment Associated with Exudative Age-Related Macular Degeneration

Rispoli

Eandi

Antonio³

et al. 2021

Biomedicines

View full text Add to dashboard Cite

Background: The purpose of this study was to describe early changes in the morphology of pigment epithelium detachments (PED) after an intravitreal injection of Brolucizumab into eyes with macular neovascularization secondary to exudative age-related macular degeneration (e-AMD). Method: We included twelve eyes of 12 patients with PED secondary to e-AMD which were not responding to prior anti-VEGF treatments. An ophthalmic examination and an assessment of PED-horizontal maximal diameter (PED-HMD), PED-maximum high (PED-MH) and macular neovascularization (MNV) flow area (MNV-FA) by the means of structural optical coherence tomography (OCT) and OCT Angiography (OCT-A) were performed at baseline, as well as 1, 7, 14 and 30 days after the injection. Results: The mean age of the population of study was 78.4 (SD ± 4.8). The mean number of previous Ranibizumab or Aflibercept injections was 13 (SD ± 8). At the last follow-up visit, the PED-HMD did not significantly change (p = 0.16; F(DF:1.94, 20,85) = 1.9), the PED-MH showed a significant reduction [p = 0.01; F(DF:1.31, 14.13) = 6.84.] and the MNV-FA did not significantly differ (p = 0.1; F(1.97, 21.67) = 2.54) from baseline. No signs of ocular inflammation were observed during follow-up. Conclusions: A single Brolucizumab injection was able to determine the short-term effects on PEDs’ anatomical features of eyes with an unresponsive e-AMD.

show abstract

Section: Discussionmentioning

confidence: 99%

Biomarkers in Early Response to Brolucizumab on Pigment Epithelium Detachment Associated with Exudative Age-Related Macular Degeneration

Rispoli

Eandi

Antonio³

et al. 2021

Biomedicines

View full text Add to dashboard Cite

show abstract

“…As an example, eyes with diabetic macular edema are frequently characterized by segmentation errors, as recently demonstrated by Ghasemi Falavarjani and colleagues [16]. A recent study using convolutional neural networks (CNNs) has demonstrated the utility of this algorithm to improve segmentation of OCTA data [17].…”

Section: Limitations and Artifactsmentioning

confidence: 94%

Guidelines on Optical Coherence Tomography Angiography Imaging: 2020 Focused Update

et al. 2020

View full text Add to dashboard Cite

Optical coherence tomography angiography (OCTA) has significantly expanded our knowledge of the ocular vasculature. In this review, we provide a discussion of the fundamental principles of OCTA and the application of this imaging modality to study the retinal and choroidal vessels. These guidelines are focused on 2020, and include updates since the 2019 publication. Importantly, we will comment on recent findings on OCTA technology with a special focus on the three-dimensional (3D) OCTA visualization.

show abstract

“…Many recent papers dealt with the contributions of artificial intelligence (AI) and deep learning (DL) approaches in ophthalmology [13][14][15]. However, only few works have addressed CNV screening on OCTA images using DL methods [16]. Most of the existing works focused on the diagnosis of AMD using OCT or Color Fundus Photography.…”

Section: Cnn Architecture and Transfer Learningmentioning

confidence: 99%

“…The obtained results showed an accuracy of 87.27% in differentiating healthy, no DR and DR eyes. In the same period, Wang et al [16] developed an algorithm based on two CNNs to classify input OCTA images (using structural volumes and enface retinal angiograms) as CNV or Non-CNV and then segment the CNV membrane when present. The proposed neural network included a cutoff threshold for CNV area to overcome the residual artifacts limitation that could be confounded with CNV.…”

Section: Cnn Architecture and Transfer Learningmentioning

confidence: 99%

Choroidal Neovascularization Screening on OCT-Angiography Choriocapillaris Images by Convolutional Neural Networks

et al. 2021

View full text Add to dashboard Cite

Choroidal Neovascularization (CNV) is the advanced stage of Age-related Macular Degeneration (AMD), which is the leading cause of irreversible visual loss for elder people in developed countries. Optical Coherence Tomography Angiography (OCTA) is a recent non-invasive imaging technique widely used nowadays in diagnosis and follow-up of CNV. In this study, an automatic screening of CNV based on deep learning is performed using OCTA choriocapillaris images. CNV eyes (advanced wet AMD) are diagnosed among healthy eyes (no AMD) and eyes with drusen (intermediate AMD). An OCTA dataset of 1396 images is used to train and evaluate the model. A pre-trained convolutional neural network (CNN) is fine-tuned and validated on 80% of the dataset while the remaining 20% is used independently for predictions. The model can accurately detect CNV on the test set with an accuracy of 89.74%, precision of 0.96 and 0.99 area under the curve of the receiver operating characteristic. A good overall classification accuracy of 88.46% is obtained on a balanced test set. Detailed analysis of misclassified images shows that they are also considered ambiguous images for expert clinicians. This novel CNN-based application is truly a breakthrough to assist clinicians in the challenging task of screening for neovascular complications.

show abstract

Automated diagnosis and segmentation of choroidal neovascularization in OCT angiography using deep learning

Cited by 59 publications

References 49 publications

Biomarkers in Early Response to Brolucizumab on Pigment Epithelium Detachment Associated with Exudative Age-Related Macular Degeneration

Biomarkers in Early Response to Brolucizumab on Pigment Epithelium Detachment Associated with Exudative Age-Related Macular Degeneration

Guidelines on Optical Coherence Tomography Angiography Imaging: 2020 Focused Update

Choroidal Neovascularization Screening on OCT-Angiography Choriocapillaris Images by Convolutional Neural Networks

Contact Info

Product

Resources

About