Artificial Intelligence in Eye Disease

A special issue of Diagnostics (ISSN 2075-4418). This special issue belongs to the section "Machine Learning and Artificial Intelligence in Diagnostics".

Deadline for manuscript submissions: closed (31 May 2022) | Viewed by 43219

Special Issue Editor

1. Department of Brain and Cognitive Engineering, Korea University, Seoul 136-701, Republic of Korea
2. Department of Artificial Intelligence, Korea University, Seoul 136-701, Republic of Korea
Interests: artificial intelligence in biomedicine; diagnosis of retinal diseases; deep learning for ophthalmology images; neuroscience research
Special Issues, Collections and Topics in MDPI journals

Special Issue Information

Dear Colleagues,

While the use of artificial intelligence (AI) is rapidly spreading to the medical world amid the vortex of the fourth industrial revolution, the use of AI in ophthalmology is attracting attention for diagnosis of various ophthalmic diseases, including optic nerve diseases, which are difficult to diagnose. Particularly, it could help to diagnose with high accuracy by introducing the AI when applied to fundus photographs, optical coherence tomography, and visual field to achieve strong classification performance in the detection of ocular and retinal diseases. In ocular imaging, AI can be used as a possible solution for screening, diagnosing, and monitoring patients with major eye disease in primary care and community settings. For instance, through deep learning algorithms that read retinal images, various diseases can be observed, such as bleeding, macular abnormalities—e.g., drusen—choroidal abnormalities, retinal vessel abnormalities, nerve fiber layer defects, and glaucomatous optic nerve papilla changes. Thus, deep learning architecture can be applied to learn to recognize eye diseases, thereby raising the diagnosis rate with a clinically acceptable performance. In other words, AI serves as a safety device for both patients and doctors, and as an auxiliary tool to quickly judge the results. It prevents the possibility of misdiagnosis that can occur in the first place, provides treatment efficiency, and increases patient reliability. Consequently, AI could potentially revolutionize the way that ophthalmology is practiced in the future. Thus, the aims of this Special Issue are to highlight the recent progress and trends in utilizing AI techniques, such as machine learning and deep learning for detecting, screening, diagnosing, and monitoring numerous eye diseases not only in diverse clinical practice but also in basic research of ophthalmology.

Prof. Dr. Jae-Ho Han
Guest Editor

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Diagnostics is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2600 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

  • Artificial intelligence
  •  Deep learning
  •  Fundus image
  •  Optical coherence tomography
  •  Ophthalmology
  •  Retinal vessel
  •  Glaucoma
  •  Retinopathy
  •  Macular degeneration
  •  Image segmentation.

Related Special Issues

Published Papers (12 papers)

Order results
Result details
Select all
Export citation of selected articles as:

Editorial

Jump to: Research, Review

4 pages, 198 KiB  
Editorial
Artificial Intelligence in Eye Disease: Recent Developments, Applications, and Surveys
Diagnostics 2022, 12(8), 1927; https://doi.org/10.3390/diagnostics12081927 - 10 Aug 2022
Cited by 5 | Viewed by 1837
Abstract
Artificial intelligence (AI) has expanded by finding applications in medical diagnosis for clinical support systems [...] Full article
(This article belongs to the Special Issue Artificial Intelligence in Eye Disease)

Research

Jump to: Editorial, Review

14 pages, 5445 KiB  
Article
Improving Glaucoma Diagnosis Assembling Deep Networks and Voting Schemes
Diagnostics 2022, 12(6), 1382; https://doi.org/10.3390/diagnostics12061382 - 02 Jun 2022
Cited by 6 | Viewed by 4034
Abstract
Glaucoma is a group of eye conditions that damage the optic nerve, the health of which is vital for good eyesight. This damage is often caused by higher-than-normal pressure in the eye. In the past few years, the applications of artificial intelligence and [...] Read more.
Glaucoma is a group of eye conditions that damage the optic nerve, the health of which is vital for good eyesight. This damage is often caused by higher-than-normal pressure in the eye. In the past few years, the applications of artificial intelligence and data science have increased rapidly in medicine especially in imaging applications. In particular, deep learning tools have been successfully applied obtaining, in some cases, results superior to those obtained by humans. In this article, we present a soft novel ensemble model based on the K-NN algorithm, that combines the probability of class membership obtained by several deep learning models. In this research, three models of different nature (CNN, CapsNets and Convolutional Autoencoders) have been selected searching for diversity. The latent space of these models are combined using the local information provided by the true sample labels and the K-NN algorithm is applied to determine the final decision. The results obtained on two different datasets of retinal images show that the proposed ensemble model improves the diagnosis capabilities for both the individual models and the state-of-the-art results. Full article
(This article belongs to the Special Issue Artificial Intelligence in Eye Disease)
Show Figures

Figure 1

23 pages, 11027 KiB  
Article
Intelligent Diagnosis and Classification of Keratitis
Diagnostics 2022, 12(6), 1344; https://doi.org/10.3390/diagnostics12061344 - 28 May 2022
Cited by 13 | Viewed by 2380
Abstract
A corneal ulcer is an open sore that forms on the cornea; it is usually caused by an infection or injury and can result in ocular morbidity. Early detection and discrimination between different ulcer diseases reduces the chances of visual disability. Traditional clinical [...] Read more.
A corneal ulcer is an open sore that forms on the cornea; it is usually caused by an infection or injury and can result in ocular morbidity. Early detection and discrimination between different ulcer diseases reduces the chances of visual disability. Traditional clinical methods that use slit-lamp images can be tiresome, expensive, and time-consuming. Instead, this paper proposes a deep learning approach to diagnose corneal ulcers, enabling better, improved treatment. This paper suggests two modes to classify corneal images using manual and automatic deep learning feature extraction. Different dimensionality reduction techniques are utilized to uncover the most significant features that give the best results. Experimental results show that manual and automatic feature extraction techniques succeeded in discriminating ulcers from a general grading perspective, with ~93% accuracy using the 30 most significant features extracted using various dimensionality reduction techniques. On the other hand, automatic deep learning feature extraction discriminated severity grading with a higher accuracy than type grading regardless of the number of features used. To the best of our knowledge, this is the first report to ever attempt to distinguish corneal ulcers based on their grade grading, type grading, ulcer shape, and distribution. Identifying corneal ulcers at an early stage is a preventive measure that reduces aggravation and helps track the efficacy of adapted medical treatment, improving the general public health in remote, underserved areas. Full article
(This article belongs to the Special Issue Artificial Intelligence in Eye Disease)
Show Figures

Figure 1

26 pages, 7149 KiB  
Article
Identifying Those at Risk of Glaucoma: A Deep Learning Approach for Optic Disc and Cup Segmentation and Their Boundary Analysis
Diagnostics 2022, 12(5), 1063; https://doi.org/10.3390/diagnostics12051063 - 24 Apr 2022
Cited by 8 | Viewed by 4969
Abstract
Glaucoma is a leading cause of irreversible vision loss that gradually damages the optic nerve. In ophthalmic fundus images, measurements of the cup to optic disc (CD) ratio, CD area ratio, neuroretinal rim to optic disc (RD) area ratio, and rim thickness are [...] Read more.
Glaucoma is a leading cause of irreversible vision loss that gradually damages the optic nerve. In ophthalmic fundus images, measurements of the cup to optic disc (CD) ratio, CD area ratio, neuroretinal rim to optic disc (RD) area ratio, and rim thickness are key measures to screen for potential glaucomatous damage. We propose an automatic method using deep learning algorithms to segment the optic disc and cup and to estimate the key measures. The proposed method comprises three steps: The Region of Interest (ROI) (location of the optic disc) detection from a fundus image using Mask R-CNN, the optic disc and cup segmentation from the ROI using the proposed Multiscale Average Pooling Net (MAPNet), and the estimation of the key measures. Our segmentation results using 1099 fundus images show 0.9381 Jaccard Index (JI) and 0.9679 Dice Coefficient (DC) for the optic disc and 0.8222 JI and 0.8996 DC for the cup. The average CD, CD area, and RD ratio errors are 0.0451, 0.0376, and 0.0376, respectively. The average disc, cup, and rim radius ratio errors are 0.0500, 0.2257, and 0.2166, respectively. Our method performs well in estimating the key measures and shows potential to work within clinical pathways once fully implemented. Full article
(This article belongs to the Special Issue Artificial Intelligence in Eye Disease)
Show Figures

Figure 1

13 pages, 3420 KiB  
Article
Automatic Detection of Age-Related Macular Degeneration Based on Deep Learning and Local Outlier Factor Algorithm
Diagnostics 2022, 12(2), 532; https://doi.org/10.3390/diagnostics12020532 - 18 Feb 2022
Cited by 12 | Viewed by 5543
Abstract
Age-related macular degeneration (AMD) is a retinal disorder affecting the elderly, and society’s aging population means that the disease is becoming increasingly prevalent. The vision in patients with early AMD is usually unaffected or nearly normal but central vision may be weakened or [...] Read more.
Age-related macular degeneration (AMD) is a retinal disorder affecting the elderly, and society’s aging population means that the disease is becoming increasingly prevalent. The vision in patients with early AMD is usually unaffected or nearly normal but central vision may be weakened or even lost if timely treatment is not performed. Therefore, early diagnosis is particularly important to prevent the further exacerbation of AMD. This paper proposed a novel automatic detection method of AMD from optical coherence tomography (OCT) images based on deep learning and a local outlier factor (LOF) algorithm. A ResNet-50 model with L2-constrained softmax loss was retrained to extract features from OCT images and the LOF algorithm was used as the classifier. The proposed method was trained on the UCSD dataset and tested on both the UCSD dataset and Duke dataset, with an accuracy of 99.87% and 97.56%, respectively. Even though the model was only trained on the UCSD dataset, it obtained good detection accuracy when tested on another dataset. Comparison with other methods also indicates the efficiency of the proposed method in detecting AMD. Full article
(This article belongs to the Special Issue Artificial Intelligence in Eye Disease)
Show Figures

Figure 1

14 pages, 2321 KiB  
Article
Deep Learning Prediction of Response to Anti-VEGF among Diabetic Macular Edema Patients: Treatment Response Analyzer System (TRAS)
Diagnostics 2022, 12(2), 312; https://doi.org/10.3390/diagnostics12020312 - 26 Jan 2022
Cited by 13 | Viewed by 3476
Abstract
Diabetic macular edema (DME) is the most common cause of visual impairment among patients with diabetes mellitus. Anti-vascular endothelial growth factors (Anti-VEGFs) are considered the first line in its management. The aim of this research has been to develop a deep learning (DL) [...] Read more.
Diabetic macular edema (DME) is the most common cause of visual impairment among patients with diabetes mellitus. Anti-vascular endothelial growth factors (Anti-VEGFs) are considered the first line in its management. The aim of this research has been to develop a deep learning (DL) model for predicting response to intravitreal anti-VEGF injections among DME patients. The research included treatment naive DME patients who were treated with anti-VEGF. Patient’s pre-treatment and post-treatment clinical and macular optical coherence tomography (OCT) were assessed by retina specialists, who annotated pre-treatment images for five prognostic features. Patients were also classified based on their response to treatment in their post-treatment OCT into either good responder, defined as a reduction of thickness by >25% or 50 µm by 3 months, or poor responder. A novel modified U-net DL model for image segmentation, and another DL EfficientNet-B3 model for response classification were developed and implemented for predicting response to anti-VEGF injections among patients with DME. Finally, the classification DL model was compared with different levels of ophthalmology residents and specialists regarding response classification accuracy. The segmentation deep learning model resulted in segmentation accuracy of 95.9%, with a specificity of 98.9%, and a sensitivity of 87.9%. The classification accuracy of classifying patients’ images into good and poor responders reached 75%. Upon comparing the model’s performance with practicing ophthalmology residents, ophthalmologists and retina specialists, the model’s accuracy is comparable to ophthalmologist’s accuracy. The developed DL models can segment and predict response to anti-VEGF treatment among DME patients with comparable accuracy to general ophthalmologists. Further training on a larger dataset is nonetheless needed to yield more accurate response predictions. Full article
(This article belongs to the Special Issue Artificial Intelligence in Eye Disease)
Show Figures

Figure 1

14 pages, 2057 KiB  
Article
Interleaving Automatic Segmentation and Expert Opinion for Retinal Conditions
Diagnostics 2022, 12(1), 22; https://doi.org/10.3390/diagnostics12010022 - 23 Dec 2021
Cited by 7 | Viewed by 2168
Abstract
Optical coherence tomography (OCT) has become the leading diagnostic tool in modern ophthalmology. We are interested here in developing a support tool for the segmentation of retina layers. The proposed method relies on graph theory and geodesic distance. As each retina layer is [...] Read more.
Optical coherence tomography (OCT) has become the leading diagnostic tool in modern ophthalmology. We are interested here in developing a support tool for the segmentation of retina layers. The proposed method relies on graph theory and geodesic distance. As each retina layer is characterised by different features, the proposed method interleaves various gradients during detection, such as horizontal and vertical gradients or open-closed gradients. The method was tested on a dataset of 750 OCT B-Scan Spectralis provided by the Ophthalmology Department of the County Emergency Hospital Cluj-Napoca. The method has smaller signed error on layers B1, B7 and B8, with the highest value of 0.43 pixels. The average value of signed error on all layers is −1.99 ± 1.14 px. The average value for mean absolute error is 2.60 ± 0.95 px. Since the target is a support tool for the human agent, the ophthalmologist can intervene after each automatic step. Human intervention includes validation or fine tuning of the automatic segmentation. In line with design criteria advocated by explainable artificial intelligence (XAI) and human-centered AI, this approach gives more control and transparency as well as more of a global perspective on the segmentation process. Full article
(This article belongs to the Special Issue Artificial Intelligence in Eye Disease)
Show Figures

Figure 1

11 pages, 248 KiB  
Article
Testing a Deep Learning Algorithm for Detection of Diabetic Retinopathy in a Spanish Diabetic Population and with MESSIDOR Database
Diagnostics 2021, 11(8), 1385; https://doi.org/10.3390/diagnostics11081385 - 31 Jul 2021
Cited by 18 | Viewed by 2176
Abstract
Background: The aim of the present study was to test our deep learning algorithm (DLA) by reading the retinographies. Methods: We tested our DLA built on convolutional neural networks in 14,186 retinographies from our population and 1200 images extracted from MESSIDOR. The retinal [...] Read more.
Background: The aim of the present study was to test our deep learning algorithm (DLA) by reading the retinographies. Methods: We tested our DLA built on convolutional neural networks in 14,186 retinographies from our population and 1200 images extracted from MESSIDOR. The retinal images were graded both by the DLA and independently by four retina specialists. Results of the DLA were compared according to accuracy (ACC), sensitivity (S), specificity (SP), positive predictive value (PPV), negative predictive value (NPV), and area under the receiver operating characteristic curve (AUC), distinguishing between identification of any type of DR (any DR) and referable DR (RDR). Results: The results of testing the DLA for identifying any DR in our population were: ACC = 99.75, S = 97.92, SP = 99.91, PPV = 98.92, NPV = 99.82, and AUC = 0.983. When detecting RDR, the results were: ACC = 99.66, S = 96.7, SP = 99.92, PPV = 99.07, NPV = 99.71, and AUC = 0.988. The results of testing the DLA for identifying any DR with MESSIDOR were: ACC = 94.79, S = 97.32, SP = 94.57, PPV = 60.93, NPV = 99.75, and AUC = 0.959. When detecting RDR, the results were: ACC = 98.78, S = 94.64, SP = 99.14, PPV = 90.54, NPV = 99.53, and AUC = 0.968. Conclusions: Our DLA performed well, both in detecting any DR and in classifying those eyes with RDR in a sample of retinographies of type 2 DM patients in our population and the MESSIDOR database. Full article
(This article belongs to the Special Issue Artificial Intelligence in Eye Disease)
11 pages, 2247 KiB  
Article
Using Slit-Lamp Images for Deep Learning-Based Identification of Bacterial and Fungal Keratitis: Model Development and Validation with Different Convolutional Neural Networks
Diagnostics 2021, 11(7), 1246; https://doi.org/10.3390/diagnostics11071246 - 12 Jul 2021
Cited by 36 | Viewed by 3035
Abstract
In this study, we aimed to develop a deep learning model for identifying bacterial keratitis (BK) and fungal keratitis (FK) by using slit-lamp images. We retrospectively collected slit-lamp images of patients with culture-proven microbial keratitis between 1 January 2010 and 31 December 2019 [...] Read more.
In this study, we aimed to develop a deep learning model for identifying bacterial keratitis (BK) and fungal keratitis (FK) by using slit-lamp images. We retrospectively collected slit-lamp images of patients with culture-proven microbial keratitis between 1 January 2010 and 31 December 2019 from two medical centers in Taiwan. We constructed a deep learning algorithm consisting of a segmentation model for cropping cornea images and a classification model that applies different convolutional neural networks (CNNs) to differentiate between FK and BK. The CNNs included DenseNet121, DenseNet161, DenseNet169, DenseNet201, EfficientNetB3, InceptionV3, ResNet101, and ResNet50. The model performance was evaluated and presented as the area under the curve (AUC) of the receiver operating characteristic curves. A gradient-weighted class activation mapping technique was used to plot the heat map of the model. By using 1330 images from 580 patients, the deep learning algorithm achieved the highest average accuracy of 80.0%. Using different CNNs, the diagnostic accuracy for BK ranged from 79.6% to 95.9%, and that for FK ranged from 26.3% to 65.8%. The CNN of DenseNet161 showed the best model performance, with an AUC of 0.85 for both BK and FK. The heat maps revealed that the model was able to identify the corneal infiltrations. The model showed a better diagnostic accuracy than the previously reported diagnostic performance of both general ophthalmologists and corneal specialists. Full article
(This article belongs to the Special Issue Artificial Intelligence in Eye Disease)
Show Figures

Figure 1

16 pages, 1358 KiB  
Article
Group and Shuffle Convolutional Neural Networks with Pyramid Pooling Module for Automated Pterygium Segmentation
Diagnostics 2021, 11(6), 1104; https://doi.org/10.3390/diagnostics11061104 - 17 Jun 2021
Cited by 12 | Viewed by 2011
Abstract
Pterygium is an eye condition that is prevalent among workers that are frequently exposed to sunlight radiation. However, most of them are not aware of this condition, which motivates many volunteers to set up health awareness booths to give them free health screening. [...] Read more.
Pterygium is an eye condition that is prevalent among workers that are frequently exposed to sunlight radiation. However, most of them are not aware of this condition, which motivates many volunteers to set up health awareness booths to give them free health screening. As a result, a screening tool that can be operated on various platforms is needed to support the automated pterygium assessment. One of the crucial functions of this assessment is to extract the infected regions, which directly correlates with the severity levels. Hence, Group-PPM-Net is proposed by integrating a spatial pyramid pooling module (PPM) and group convolution to the deep learning segmentation network. The system uses a standard mobile phone camera input, which is then fed to a modified encoder-decoder convolutional neural network, inspired by a Fully Convolutional Dense Network that consists of a total of 11 dense blocks. A PPM is integrated into the network because of its multi-scale capability, which is useful for multi-scale tissue extraction. The shape of the tissues remains relatively constant, but the size will differ according to the severity levels. Moreover, group and shuffle convolution modules are also integrated at the decoder side of Group-PPM-Net by placing them at the starting layer of each dense block. The addition of these modules allows better correlation among the filters in each group, while the shuffle process increases channel variation that the filters can learn from. The results show that the proposed method obtains mean accuracy, mean intersection over union, Hausdorff distance, and Jaccard index performances of 0.9330, 0.8640, 11.5474, and 0.7966, respectively. Full article
(This article belongs to the Special Issue Artificial Intelligence in Eye Disease)
Show Figures

Figure 1

Review

Jump to: Editorial, Research

18 pages, 900 KiB  
Review
Computer-Assisted Pterygium Screening System: A Review
Diagnostics 2022, 12(3), 639; https://doi.org/10.3390/diagnostics12030639 - 05 Mar 2022
Cited by 7 | Viewed by 4604
Abstract
Pterygium is an eye condition that causes the fibrovascular tissues to grow towards the corneal region. At the early stage, it is not a harmful condition, except for slight discomfort for the patients. However, it will start to affect the eyesight of the [...] Read more.
Pterygium is an eye condition that causes the fibrovascular tissues to grow towards the corneal region. At the early stage, it is not a harmful condition, except for slight discomfort for the patients. However, it will start to affect the eyesight of the patient once the tissues encroach towards the corneal region, with a more serious impact if it has grown into the pupil region. Therefore, this condition needs to be identified as early as possible to halt its growth, with the use of simple eye drops and sunglasses. One of the associated risk factors for this condition is a low educational level, which explains the reason that the majority of the patients are not aware of this condition. Hence, it is important to develop an automated pterygium screening system based on simple imaging modalities such as a mobile phone camera so that it can be assessed by many people. During the early stage of automated pterygium screening system development, conventional machine learning techniques such as support vector machines and artificial neural networks are the de facto algorithms to detect the presence of pterygium tissues. However, with the arrival of the deep learning era, coupled with the availability of large training data, deep learning networks have replaced the conventional networks in screening for the pterygium condition. The deep learning networks have been successfully implemented for three major purposes, which are to classify an image regarding whether there is the presence of pterygium tissues or not, to localize the lesion tissues through object detection methodology, and to semantically segment the lesion tissues at the pixel level. This review paper summarizes the type, severity, risk factors, and existing state-of-the-art technology in automated pterygium screening systems. A few available datasets are also discussed in this paper for both classification and segmentation tasks. In conclusion, a computer-assisted pterygium screening system will benefit many people all over the world, especially in alerting them to the possibility of having this condition so that preventive actions can be advised at an early stage. Full article
(This article belongs to the Special Issue Artificial Intelligence in Eye Disease)
Show Figures

Figure 1

27 pages, 5113 KiB  
Review
Review of Machine Learning Applications Using Retinal Fundus Images
Diagnostics 2022, 12(1), 134; https://doi.org/10.3390/diagnostics12010134 - 06 Jan 2022
Cited by 14 | Viewed by 4341
Abstract
Automating screening and diagnosis in the medical field saves time and reduces the chances of misdiagnosis while saving on labor and cost for physicians. With the feasibility and development of deep learning methods, machines are now able to interpret complex features in medical [...] Read more.
Automating screening and diagnosis in the medical field saves time and reduces the chances of misdiagnosis while saving on labor and cost for physicians. With the feasibility and development of deep learning methods, machines are now able to interpret complex features in medical data, which leads to rapid advancements in automation. Such efforts have been made in ophthalmology to analyze retinal images and build frameworks based on analysis for the identification of retinopathy and the assessment of its severity. This paper reviews recent state-of-the-art works utilizing the color fundus image taken from one of the imaging modalities used in ophthalmology. Specifically, the deep learning methods of automated screening and diagnosis for diabetic retinopathy (DR), age-related macular degeneration (AMD), and glaucoma are investigated. In addition, the machine learning techniques applied to the retinal vasculature extraction from the fundus image are covered. The challenges in developing these systems are also discussed. Full article
(This article belongs to the Special Issue Artificial Intelligence in Eye Disease)
Show Figures

Figure 1

Back to TopTop