Detecting Anomalies in Retinal Diseases Using Generative, Discriminative, and Self-supervised Deep Learning

Background Retinal vascular diseases, including diabetic macular edema (DME), neovascular age-related macular degeneration (nAMD), myopic choroidal neovascularization (mCNV), and branch and central retinal vein occlusion (BRVO/CRVO), are considered vision-threatening eye diseases. However, accurate diagnosis depends on multimodal imaging and the expertise of retinal ophthalmologists. Objective The aim of this study was to develop a deep learning model to detect treatment-requiring retinal vascular diseases using multimodal imaging. Methods This retrospective study enrolled participants with multimodal ophthalmic imaging data from 3 hospitals in Taiwan from 2013 to 2019. Eye-related images were used, including those obtained through retinal fundus photography, optical coherence tomography (OCT), and fluorescein angiography with or without indocyanine green angiography (FA/ICGA). A deep learning model was constructed for detecting DME, nAMD, mCNV, BRVO, and CRVO and identifying treatment-requiring diseases. Model performance was evaluated and is presented as the area under the curve (AUC) for each receiver operating characteristic curve. Results A total of 2992 eyes of 2185 patients were studied, with 239, 1209, 1008, 211, 189, and 136 eyes in the control, DME, nAMD, mCNV, BRVO, and CRVO groups, respectively. Among them, 1898 eyes required treatment. The eyes were divided into training, validation, and testing groups in a 5:1:1 ratio. In total, 5117 retinal fundus photos, 9316 OCT images, and 20,922 FA/ICGA images were used. The AUCs for detecting mCNV, DME, nAMD, BRVO, and CRVO were 0.996, 0.995, 0.990, 0.959, and 0.988, respectively. The AUC for detecting treatment-requiring diseases was 0.969. From the heat maps, we observed that the model could identify retinal vascular diseases. Conclusions Our study developed a deep learning model to detect retinal diseases using multimodal ophthalmic imaging. Furthermore, the model demonstrated good performance in detecting treatment-requiring retinal diseases.

Download Full-text

Deep learning for the segmentation of preserved photoreceptors on en face optical coherence tomography in two inherited retinal diseases

Biomedical Optics Express ◽

10.1364/boe.9.003092 ◽

2018 ◽

Vol 9 (7) ◽

pp. 3092 ◽

Cited By ~ 10

Author(s):

Acner Camino ◽

Zhuo Wang ◽

Jie Wang ◽

Mark E. Pennesi ◽

Paul Yang ◽

...

Keyword(s):

Optical Coherence Tomography ◽

Deep Learning ◽

Retinal Diseases ◽

Optical Coherence ◽

En Face

Download Full-text

Deep learning for identifying and classifying retinal diseases

2020 2nd International Conference on Computer and Information Sciences (ICCIS) ◽

10.1109/iccis49240.2020.9257674 ◽

2020 ◽

Author(s):

Mohamed Berrimi ◽

Abdelouaheb Moussaoui

Keyword(s):

Deep Learning ◽

Retinal Diseases

Download Full-text

Fully Leveraging Deep Learning Methods for Constructing Retinal Fundus Photomontages

Applied Sciences ◽

10.3390/app11041754 ◽

2021 ◽

Vol 11 (4) ◽

pp. 1754

Author(s):

Jooyoung Kim ◽

Sojung Go ◽

Kyoungjin Noh ◽

Sangjun Park ◽

Soochahn Lee

Keyword(s):

Deep Learning ◽

Image Registration ◽

Experimental Evaluation ◽

Image Texture ◽

Peripheral Retina ◽

Image Artifacts ◽

Retinal Diseases ◽

Fundus Images ◽

Learning Methods ◽

Retinal Fundus

Retinal photomontages, which are constructed by aligning and integrating multiple fundus images, are useful in diagnosing retinal diseases affecting peripheral retina. We present a novel framework for constructing retinal photomontages that fully leverage recent deep learning methods. Deep learning based object detection is used to define the order of image registration and blending. Deep learning based vessel segmentation is used to enhance image texture to improve registration performance within a two step image registration framework comprising rigid and non-rigid registration. Experimental evaluation demonstrates the robustness of our montage construction method with an increased amount of successfully integrated images as well as reduction of image artifacts.

Download Full-text

Deep learning-based automated detection of retinal diseases using optical coherence tomography images

Biomedical Optics Express ◽

10.1364/boe.10.006204 ◽

2019 ◽

Vol 10 (12) ◽

pp. 6204 ◽

Cited By ~ 8

Author(s):

Feng Li ◽

Hua Chen ◽

Zheng Liu ◽

Xue-dian Zhang ◽

Min-shan Jiang ◽

...

Keyword(s):

Optical Coherence Tomography ◽

Deep Learning ◽

Automated Detection ◽

Retinal Diseases ◽

Optical Coherence

Download Full-text

A Multimodal Imaging–Based Deep Learning Model for Detecting Treatment-Requiring Retinal Vascular Diseases: Model Development and Validation Study (Preprint)

10.2196/preprints.28868 ◽

2021 ◽

Author(s):

Eugene Yu-Chuan Kang ◽

Ling Yeung ◽

Yi-Lun Lee ◽

Cheng-Hsiu Wu ◽

Shu-Yen Peng ◽

...

Keyword(s):

Deep Learning ◽

Multimodal Imaging ◽

Vascular Diseases ◽

Learning Model ◽

Age Related Macular Degeneration ◽

Fundus Photography ◽

Retinal Diseases ◽

Ophthalmic Imaging ◽

Retinal Fundus ◽

Deep Learning Model

BACKGROUND Retinal vascular diseases, including diabetic macular edema (DME), neovascular age-related macular degeneration (nAMD), myopic choroidal neovascularization (mCNV), and branch and central retinal vein occlusion (BRVO/CRVO), are considered vision-threatening eye diseases. However, accurate diagnosis depends on multimodal imaging and the expertise of retinal ophthalmologists. OBJECTIVE The aim of this study was to develop a deep learning model to detect treatment-requiring retinal vascular diseases using multimodal imaging. METHODS This retrospective study enrolled participants with multimodal ophthalmic imaging data from 3 hospitals in Taiwan from 2013 to 2019. Eye-related images were used, including those obtained through retinal fundus photography, optical coherence tomography (OCT), and fluorescein angiography with or without indocyanine green angiography (FA/ICGA). A deep learning model was constructed for detecting DME, nAMD, mCNV, BRVO, and CRVO and identifying treatment-requiring diseases. Model performance was evaluated and is presented as the area under the curve (AUC) for each receiver operating characteristic curve. RESULTS A total of 2992 eyes of 2185 patients were studied, with 239, 1209, 1008, 211, 189, and 136 eyes in the control, DME, nAMD, mCNV, BRVO, and CRVO groups, respectively. Among them, 1898 eyes required treatment. The eyes were divided into training, validation, and testing groups in a 5:1:1 ratio. In total, 5117 retinal fundus photos, 9316 OCT images, and 20,922 FA/ICGA images were used. The AUCs for detecting mCNV, DME, nAMD, BRVO, and CRVO were 0.996, 0.995, 0.990, 0.959, and 0.988, respectively. The AUC for detecting treatment-requiring diseases was 0.969. From the heat maps, we observed that the model could identify retinal vascular diseases. CONCLUSIONS Our study developed a deep learning model to detect retinal diseases using multimodal ophthalmic imaging. Furthermore, the model demonstrated good performance in detecting treatment-requiring retinal diseases.

Download Full-text

Deep Learning to Distinguish ABCA4-Related Stargardt Disease from PRPH2-Related Pseudo-Stargardt Pattern Dystrophy

Journal of Clinical Medicine ◽

10.3390/jcm10245742 ◽

2021 ◽

Vol 10 (24) ◽

pp. 5742

Author(s):

Alexandra Miere ◽

Olivia Zambrowski ◽

Arthur Kessler ◽

Carl-Joe Mehanna ◽

Carlotta Pallone ◽

...

Keyword(s):

Neural Network ◽

Deep Learning ◽

Good Accuracy ◽

Fundus Autofluorescence ◽

Retinal Diseases ◽

Proof Of Concept ◽

Stargardt Disease ◽

Test Dataset ◽

Pattern Dystrophy ◽

Learning Classifier

(1) Background: Recessive Stargardt disease (STGD1) and multifocal pattern dystrophy simulating Stargardt disease (“pseudo-Stargardt pattern dystrophy”, PSPD) share phenotypic similitudes, leading to a difficult clinical diagnosis. Our aim was to assess whether a deep learning classifier pretrained on fundus autofluorescence (FAF) images can assist in distinguishing ABCA4-related STGD1 from the PRPH2/RDS-related PSPD and to compare the performance with that of retinal specialists. (2) Methods: We trained a convolutional neural network (CNN) using 729 FAF images from normal patients or patients with inherited retinal diseases (IRDs). Transfer learning was then used to update the weights of a ResNet50V2 used to classify the 370 FAF images into STGD1 and PSPD. Retina specialists evaluated the same dataset. The performance of the CNN and that of retina specialists were compared in terms of accuracy, sensitivity, and precision. (3) Results: The CNN accuracy on the test dataset of 111 images was 0.882. The AUROC was 0.890, the precision was 0.883 and the sensitivity was 0.883. The accuracy for retina experts averaged 0.816, whereas for retina fellows it averaged 0.724. (4) Conclusions: This proof-of-concept study demonstrates that, even with small databases, a pretrained CNN is able to distinguish between STGD1 and PSPD with good accuracy.

Download Full-text