Rapid classification of glaucomatous fundus images
Vasudevan lakshminarayanan2021, Journal of the Optical Society of America
Sign up for access to the world's latest research
checkGet notified about relevant papers
checkSave papers to use in your research
checkJoin the discussion with peers
checkTrack your impact
Abstract
We propose a new method for training convolutional neural networks which integrates reinforcement learning along with supervised learning and use it for transfer learning for classification of glaucoma from colored fundus images. The training method uses hill climbing techniques via two different climber types viz. "random movement" and "random detection" integrated with supervised learning model through stochastic gradient descent with momentum (SGDM) model. The model was trained and tested using the Drishti-GS, and RIM-ONE-r2 datasets having glaucomatous and normal fundus images. The performance for prediction was tested by transfer learning on five CNN architectures namely, GoogLeNet, NASNet,. A 5-fold classification was used for evaluating the performance and high sensitivities while maintaining high accuracies were achieved. Of the models tested, DenseNet-201 architecture performed the best in terms of sensitivity and area under the curve (AUC). This method of training allows transfer learning on small datasets and can be applied for tele-ophthalmology applications including training with local datasets.
Related papers
Performance Analysis of Optimizers for Glaucoma Diagnosis from Fundus Images Using Transfer Learning
2021
Transferring the weights from the pre-trained model results in faster and easier training than training the network from scratch. The proper choice of optimizer may improve the performance of the deep neural networks for image classification problems. This paper analyzes and compares three standard first-order optimizers like stochastic gradient descent with momentum (SGDM), adaptive moment estimation (Adam), and root mean square propagation (RMSProp), particularly for detecting glaucoma from fundus images using different CNN architectures like AlexNet, VGG-19, and ResNet-101. Experiment results show that network parameters updated using Adam optimizer yields better results in most of the databases. Among the models, VGG-19 has obtained the highest classification accuracy of 91.71, 87.8, and 97.12%, in DRISHTI-GS1, RIM-ONE(2), and LAG databases, respectively. ResNet-101 has outperformed other networks in ORIGA and ACRIMA databases, with the highest classification accuracy of 80.5% and 98.5%, respectively.
Glaucoma Detection Using Transfer Learning with the Faster R-CNN Model and a ResNet-50-FPN Backbone
Proceedings of the 26th International Conference on Enterprise Information Systems (ICEIS), 2024
Early detection of glaucoma has the potential to prevent vision loss. The application of artificial intelligence can enhance the cost-effectiveness of glaucoma detection by reducing the need for manual intervention. Glau- coma is the second leading cause of blindness and, due to its asymptomatic nature until advanced stages, diagnosis is often delayed. Having a general understanding of the disease’s pathophysiology, diagnosis, and treatment can assist primary care physicians in referring high-risk patients for comprehensive ophthalmologic examinations and actively participating in the care of individuals affected by this condition. This article describes a method for glaucoma detection with the Faster R-CNN model and a ResNet-50-FPN backbone. Our experiments demonstrated greater accuracy compared to models such as, AlexNet, VGG-11, VGG-16, VGG-19, GoogleNet-V1, ResNet-18, ResNet-50, ResNet-101 and ResNet-152.
Deployment of CNN on colour fundus images for the automatic detection of glaucoma
International Journal of Applied Science and Engineering
Detection of glaucoma has become critical, as it has arisen as the subsequent essential driver of visual impairment, around the world. At present, most of the algorithms in use rely on pre-trained deep neural networks to produce the best results. However, the high computational time and complexity and the need of a large database, make glaucomadetection arduous and difficult. Keeping these in mind, this paper proposes a new convolutional neural network architecture, in particular, ProspectNet, which has demonstrated to accomplish a better accuracy with lesser computational time and complexity when tested against two pre-trained networks: VGG16 and DenseNet121. The data set is an amalgamation of two publicly available datasets-DRISHTI-GS and Glaucoma Dataset (Kaggle), comprising ocular colour fundus images of glaucomatous as well as normal eyes. ProspectNet has accomplished a normal AUC (area under the curve) as 0.991, specificity, and precision as 0.98. Confusion matrices also plotted to illustrate the new architecture's efficacy. These outcomes demonstrate that ProspectNet is a hearty option in contrast to other best in class calculations for a medium sized dataset. The paper suggests three distinct structures for glaucoma detection. One advantage of our approach is that no special feature selection, such as detailed measurements of particular traits like the structure of the optic nerve head, is necessary.
A deep learning model for the detection of both advanced and early glaucoma using fundus photography
PLOS ONE
To build a deep learning model to diagnose glaucoma using fundus photography. Design Cross sectional case study Subjects, Participants and Controls: A total of 1,542 photos (786 normal controls, 467 advanced glaucoma and 289 early glaucoma patients) were obtained by fundus photography. Method The whole dataset of 1,542 images were split into 754 training, 324 validation and 464 test datasets. These datasets were used to construct simple logistic classification and convolutional neural network using Tensorflow. The same datasets were used to fine tune pretrained GoogleNet Inception v3 model. Results The simple logistic classification model showed a training accuracy of 82.9%, validation accuracy of 79.9% and test accuracy of 77.2%. Convolutional neural network achieved accuracy and area under the receiver operating characteristic curve (AUROC) of 92.2% and 0.98 on the training data, 88.6% and 0.95 on the validation data, and 87.9% and 0.94 on the test data. Transfer-learned GoogleNet Inception v3 model achieved accuracy and AUROC of 99.7% and 0.99 on training data, 87.7% and 0.95 on validation data, and 84.5% and 0.93 on test data. Conclusion Both advanced and early glaucoma could be correctly detected via machine learning, using only fundus photographs. Our new model that is trained using convolutional neural network is more efficient for the diagnosis of early glaucoma than previously published models.
Performance of Deep Learning Architectures and Transfer Learning for Detecting Glaucomatous Optic Neuropathy in Fundus Photographs
Scientific Reports, 2018
The ability of deep learning architectures to identify glaucomatous optic neuropathy (GON) in fundus photographs was evaluated. A large database of fundus photographs (n = 14,822) from a racially and ethnically diverse group of individuals (over 33% of African descent) was evaluated by expert reviewers and classified as GON or healthy. Several deep learning architectures and the impact of transfer learning were evaluated. The best performing model achieved an overall area under receiver operating characteristic (AUC) of 0.91 in distinguishing GON eyes from healthy eyes. It also achieved an AUC of 0.97 for identifying GON eyes with moderate-to-severe functional loss and 0.89 for GON eyes with mild functional loss. A sensitivity of 88% at a set 95% specificity was achieved in detecting moderate-to-severe GON. In all cases, transfer improved performance and reduced training time. Model visualizations indicate that these deep learning models relied on, in part, anatomical features in the inferior and superior regions of the optic disc, areas commonly used by clinicians to diagnose GON. The results suggest that deep learning-based assessment of fundus images could be useful in clinical decision support systems and in the automation of large-scale glaucoma detection and screening programs. Glaucoma is characterized by progressive structural and functional damage to the optic nerve head that can eventually lead to functional impairment, disability, and blindness. It is one of the most common causes of irreversible blindness worldwide and is expected to affect roughly 80 million people by 2020 1-3. Current estimates suggest that roughly 50% of people suffering from glaucoma in the developed world are currently undiagnosed and aging populations suggests that the impact of glaucoma will continue to rise 4,5. Effective screening programs to detect glaucoma are needed to address this growing public health problem 6,7. Despite extensive research on glaucoma screening, the U.S. Preventive Services Task Force does not recommend population screening for glaucoma at least in part due the relatively low sensitivity and specificity of glaucoma screening tests given the low prevalence of the disease and insufficient evidence that the benefits of screening outweigh the costs and potential harm 8. Traditionally, clinicians have examined the ONH with ophthalmoscopy and fundus photography to diagnose and monitor glaucoma. Subjective qualitative ONH evaluation requires significant clinical training and agreement is limited even among subspecialty-trained clinicians 9-11. Automated methods that use techniques from artificial intelligence to objectively interpret images of the ONH and surrounding fundus can help address this issue. These automated methods may also be used as part of decision support systems in clinical management of glaucoma through incorporation into fundus cameras, electronic medical record systems, or picture archiving and communication systems (PACS). If these automated systems can sufficiently improve sensitivity and specificity beyond previously studied glaucoma-associated measurements, they may pave the way for large-scale screening programs. The Food and Drug Administration (FDA) has recently approved the use of an automated system
Automated Glaucoma Diagnosis Using Deep and Transfer Learning: Proposal of a System for Clinical Testing
2018 International Conference on Image and Vision Computing New Zealand (IVCNZ), 2018
We developed a deep learning algorithm for identifying glaucoma on optic nerve head (ONH) photographs. We applied transfer learning to overcome overfitting on the small training sample size that we employed. The transfer learning framework that was previously trained on large datasets such as ImageNet, uses the initial parameters and makes the approach applicable to small sample sizes. We then classified the input ONH photographs as "normal" or "glaucoma". The proposed approach achieved a validation accuracy of 92.3% on a dataset of 277 ONH photographs from normal eyes and 170 ONH photographs from eyes with glaucoma. In order to re-test the accuracy and generalizability of the proposed approach, we re-tested the algorithm using an independent dataset of 30 ONH photographs. The re-test accuracy was 80.0%.
A Review on the Detection and Classification of Glaucoma Disease Based on Transfer Learning
International journal of innovative research in computer science & technology, 2024
An eye infection is a condition affecting the eyes that can be caused by a bacterium, virus, or fungus. Numerous eye infections exist, such as glaucoma, cellulitis, keratitis, and conjunctivitis. A few of the symptoms may be itching, discharge, altered eye sight and others. Antibiotics are not effective in treating viral infections. Antibiotics treat infections caused by bacteria exclusively. A class of eye infection known as glaucoma can result in blindness and visual loss by harming the optic nerve, a nerve located at the back of the eye. You might not notice the symptoms at first because they can appear so slowly. A thorough dilated eye exam is the only way to determine if you have glaucoma. Efforts have been done to automate the procedures for the recognition and classification of glaucoma. In this paper, we have proposed a transfer learning model by reviewing pre-trained models and the model is able to provide a better accuracy. Our model is classifying the datasets into positive and negative cases during testing and validation. We utilize different prêt rained models, that are ResNet50 (90%), EffiecintNet (78%) and CNN(79%) evaluate how well they perform when trained using various optimizers. Our results show differences in accuracy and provide important information about the possibility of these models for the detection of glaucoma. An important first step towards improving the precision and dependability of glaucoma detection models in clinical settings is represented by this work.
Detection of Glaucoma on Fundus Images Using Deep Learning on a New Image Set Obtained with a Smartphone and Handheld Ophthalmoscope
Healthcare
Statistics show that an estimated 64 million people worldwide suffer from glaucoma. To aid in the detection of this disease, this paper presents a new public dataset containing eye fundus images that was developed for glaucoma pattern-recognition studies using deep learning (DL). The dataset, denoted Brazil Glaucoma, comprises 2000 images obtained from 1000 volunteers categorized into two groups: those with glaucoma (50%) and those without glaucoma (50%). All images were captured with a smartphone attached to a Welch Allyn panoptic direct ophthalmoscope. Further, a DL approach for the automatic detection of glaucoma was developed using the new dataset as input to a convolutional neural network ensemble model. The accuracy between positive and negative glaucoma detection, sensitivity, and specificity were calculated using five-fold cross-validation to train and refine the classification model. The results showed that the proposed method can identify glaucoma from eye fundus images wi...
Investigation of the Role of Convolutional Neural Network Architectures in the Diagnosis of Glaucoma using Color Fundus Photography
Turkish Journal of Ophthalmology
Objectives: To evaluate the performance of convolutional neural network (CNN) architectures to distinguish eyes with glaucoma from normal eyes. Materials and Methods: A total of 9,950 fundus photographs of 5,388 patients from the database of Eskişehir Osmangazi University Faculty of Medicine Ophthalmology Clinic were labelled as glaucoma, glaucoma suspect, or normal by three different experienced ophthalmologists. The categorized fundus photographs were evaluated using a state-of-the-art two-dimensional CNN and compared with deep residual networks (ResNet) and very deep neural networks (VGG). The accuracy, sensitivity, and specificity of glaucoma detection with the different algorithms were evaluated using a dataset of 238 normal and 320 glaucomatous fundus photographs. For the detection of suspected glaucoma, ResNet-101 architectures were tested with a data set of 170 normal, 170 glaucoma, and 167 glaucoma-suspect fundus photographs.
Detection of glaucoma from fundus image using pre-trained Densenet201 model
IJRSP Vol.50(1) [March 2021], 2021
In recent years, the performance of deep learning algorithms for image recognition has improved tremendously. The inherent ability of a convolutional neural network has made the task of classifying glaucoma and normal fundus images more appropriately. Transferring the weights from the pre-trained model resulted in faster and easier training than training the network from scratch. In this paper, a dense convolutional neural network (Densenet201) has been utilized to extract the relevant features for classification. Training with 80% of the images and testing with 20% of the images has been performed. The performance metrics obtained by various classifiers such as softmax, support vector machine (SVM), knearest neighbor (KNN), and Naive Bayes (NB) have been compared. Experimental results have shown that the softmax classifier outperformed the other classifiers with 96.48% accuracy, 98.88% sensitivity, 92.1% specificity, 95.82% precision, and 97.28% F1-score, with DRISHTI-GS1 database. An increase in the classification accuracy of about 1% has been achieved with enhanced fundus images.
An effective deep learning network for detecting and classifying glaucomatous eye
International Journal of Electrical and Computer Engineering (IJECE), 2023
Glaucoma is a well-known complex disease of the optic nerve that gradually damages eyesight due to the increase of intraocular pressure inside the eyes. Among two types of glaucoma, open-angle glaucoma is mostly happened by high intraocular pressure and can damage the eyes temporarily or sometimes permanently, another one is angle-closure glaucoma. Therefore, being diagnosed in the early stage is necessary to safe our vision. There are several ways to detect glaucomatous eyes like tonometry, perimetry, and gonioscopy but require time and expertise. Using deep learning approaches could be a better solution. This study focused on the recognition of open-angle affected eyes from the fundus images using deep learning techniques. The study evolved by applying VGG16, VGG19, and ResNet50 deep neural network architectures for classifying glaucoma positive and negative eyes. The experiment was executed on a public dataset collected from Kaggle; however, every model performed better after augmenting the dataset, and the accuracy was between 93% and 97.56%. Among the three models, VGG19 achieved the highest accuracy at 97.56%.
CNN Based Fundus Images Classification For Glaucoma Identification
International Conference on Advanced Research in Computing (ICARC), 2022
Glaucoma is a fatal, worldwide disease that can cause blindness after cataracts for people over 40-60 years. Statistics on glaucoma have shown that around 65 million people worldwide affect by glaucoma, and it is the second major reason for vision impairment after cataracts. This study uses three different Convolutional Neural Networks (CNNs) architectures, namely Inception-v3, Visual Geometry Group 19 (VGG19), Residual Neural Network 50 (ResNet50), to classify glaucoma subjects using eye fundus images. In addition, several data pre-processing and augmentation techniques were used to avoid overfitting and achieve high accuracy. The aim of this paper is to comparative analysis of the performance obtained from different configurations with CNN architectures and hyperparameter tuning. Among the considered deep learning models, the Inception-v3 model showed the highest accuracy of 98.52% for the ACRIMA fundus image dataset.
Convolutional neural network transfer for automated glaucoma identification
SPIE Proceedings, 2017
Most current systems for automated glaucoma detection in fundus images rely on segmentation-based features, which are known to be influenced by the underlying segmentation methods. Convolutional Neural Networks (CNNs) are powerful tools for solving image classification tasks as they are able to learn highly discriminative features from raw pixel intensities. However, their applicability to medical image analysis is limited by the non-availability of large sets of annotated data required for training. In this article we present results of analysis of the viability of using CNNs that are pre-trained from non-medical data for automated glaucoma detection. Two different CNNs, namely OverFeat and VGG-S, were applied to fundus images to generate feature vectors. Preprocessing techniques such as vessel inpainting, contrast-limited adaptive histogram equalization (CLAHE) or cropping around the optic nerve head (ONH) area were explored within this framework to evaluate the improvement in feature discrimination, combined with both 1 and 2 regularized logistic regression models. Results on the Drishti-GS1 dataset, evaluated in terms of area under the average ROC curve, suggests the viability of this approach and offer significant evidence of the importance of well-chosen image pre-processing for transfer learning when the amount of data is not sufficient for fine-tuning the network.
A Novel Hybrid Approach Based on Deep CNN to Detect Glaucoma Using Fundus Imaging
Electronics
Glaucoma is one of the eye diseases stimulated by the fluid pressure that increases in the eyes, damaging the optic nerves and causing partial or complete vision loss. As Glaucoma appears in later stages and it is a slow disease, detailed screening and detection of the retinal images is required to avoid vision forfeiture. This study aims to detect glaucoma at early stages with the help of deep learning-based feature extraction. Retinal fundus images are utilized for the training and testing of our proposed model. In the first step, images are pre-processed, before the region of interest (ROI) is extracted employing segmentation. Then, features of the optic disc (OD) are extracted from the images containing optic cup (OC) utilizing the hybrid features descriptors, i.e., convolutional neural network (CNN), local binary patterns (LBP), histogram of oriented gradients (HOG), and speeded up robust features (SURF). Moreover, low-level features are extracted using HOG, whereas texture fea...
Widen the Applicability of a Convolutional Neural-Network-Assisted Glaucoma Detection Algorithm of Limited Training Images across Different Datasets
Biomedicines
Automated glaucoma detection using deep learning may increase the diagnostic rate of glaucoma to prevent blindness, but generalizable models are currently unavailable despite the use of huge training datasets. This study aims to evaluate the performance of a convolutional neural network (CNN) classifier trained with a limited number of high-quality fundus images in detecting glaucoma and methods to improve its performance across different datasets. A CNN classifier was constructed using EfficientNet B3 and 944 images collected from one medical center (core model) and externally validated using three datasets. The performance of the core model was compared with (1) the integrated model constructed by using all training images from the four datasets and (2) the dataset-specific model built by fine-tuning the core model with training images from the external datasets. The diagnostic accuracy of the core model was 95.62% but dropped to ranges of 52.5–80.0% on the external datasets. Data...
Automated detection of Glaucoma using deep learning convolution network (G-net)
Multimedia Tools and Applications, 2019
Glaucoma is an ocular disease that is the leading cause of irreversible blindness due to an increased Intraocular pressure resulting in damage to the optic nerve of eye. A common method for diagnosing glaucoma progression is through examination of dilated pupil in the eye by expert ophthalmologist. But this approach is laborious and consumes a large amount of time, thus the issue can be resolved using automation by using the concept of machine learning. Convolution neural networks (CNN's) are well suited to resolve this class of problems as they can infer hierarchical information from the image which helps them to distinguish between glaucomic and non-glaucomic image patterns for diagnostic decisions. This paper presents an Artificially Intelligent glaucoma expert system based on segmentation of optic disc and optic cup. A Deep Learning architecture is developed with CNN working at its core for automating the detection of glaucoma. The proposed system uses two neural networks working in conjunction to segment optic cup and disc. The model was tested on 50 fundus images and achieved an accuracy of 95.8% for disc and 93% for cup segmentation.
An hybrid feature space from texture information and transfer learning for glaucoma classification
Journal of Visual Communication and Image Representation, 2019
Glaucoma is a progressive eye disease due to the increase in intraocular pressure. Accurate early detection may prevent vision loss. Most algorithms in the literature are not feasible for use in screening programs since they are not able to handle a wide diversity of images. We conducted an extensive study to determine the best set of features for image representation. Our feature extraction methodology included the following descriptors: LBP, GLCM, HOG, Tamura, GLRLM, morphology, and seven CNN architectures, that results in 30.682 features. Then, we used the gain ratio to order the features by importance and select the best set for glaucoma classification. Our tests were performed using 1675 images of DRISHTI, RIM-ONE, HRF, JSIEC, and ACRIMA databases. We concluded that a combination of the GLCM and pretrained CNN's has the potential to be used in a computer aid system for glaucoma detection. Our approach achieved an accuracy of 93.61%.
Web Applicable Computer-aided Diagnosis of Glaucoma Using Deep Learning
ArXiv, 2018
Glaucoma is a major eye disease, leading to vision loss in the absence of proper medical treatment. Current diagnosis of glaucoma is performed by ophthalmologists who are often analyzing several types of medical images generated by different types of medical equipment. Capturing and analyzing these medical images is labor-intensive and expensive. In this paper, we present a novel computational approach towards glaucoma diagnosis and localization, only making use of eye fundus images that are analyzed by state-of-the-art deep learning techniques. Specifically, our approach leverages Convolutional Neural Networks (CNNs) and Gradient-weighted Class Activation Mapping (Grad-CAM) for glaucoma diagnosis and localization, respectively. Quantitative and qualitative results, as obtained for a small-sized dataset with no segmentation ground truth, demonstrate that the proposed approach is promising, for instance achieving an accuracy of 0.91$\pm0.02$ and an ROC-AUC score of 0.94 for the diagn...
Multi-task Deep Learning for Glaucoma Detection From Color Fundus Images
2022
Glaucoma is an eye condition that, if not diagnosed in time, leads to loss of vision and blindness. While diagnosing campaigns are regularly launched, these require human experts to estimate in a limited time subtle changes in the shape of the optic disc from retinal fundus images. Automatic glaucoma detection methods are desirable to help with the screening of patients, reducing the diagnosis time by analyzing the increasing amount of images. Deep learning methods have been satisfactory to classify and segment diseases in retinal fundus images. To guarantee a good generalization, these methods require vast amounts of annotations, which can be highly problematic given that existing annotated glaucoma datasets contain at most few hundreds of samples. Furthermore, deep learning techniques are computationally greedy, which can be a problem in scenarios with limited resources. The objective of this work is to make optimal use of the few annotated data at disposal and obtain better gener...
Automated diagnosing primary open-angle glaucoma from fundus image by simulating human’s grading with deep learning
Scientific Reports, 2022
Primary open-angle glaucoma (POAG) is a leading cause of irreversible blindness worldwide. Although deep learning methods have been proposed to diagnose POAG, it remains challenging to develop a robust and explainable algorithm to automatically facilitate the downstream diagnostic tasks. In this study, we present an automated classification algorithm, GlaucomaNet, to identify POAG using variable fundus photographs from different populations and settings. GlaucomaNet consists of two convolutional neural networks to simulate the human grading process: learning the discriminative features and fusing the features for grading. We evaluated GlaucomaNet on two datasets: Ocular Hypertension Treatment Study (OHTS) participants and the Large-scale Attention-based Glaucoma (LAG) dataset. GlaucomaNet achieved the highest AUC of 0.904 and 0.997 for POAG diagnosis on OHTS and LAG datasets. An ensemble of network architectures further improved diagnostic accuracy. By simulating the human grading process, GlaucomaNet demonstrated high accuracy with increased transparency in POAG diagnosis (comprehensiveness scores of 97% and 36%). These methods also address two well-known challenges in the field: the need for increased image data diversity and relying heavily on perimetry for POAG diagnosis. These results highlight the potential of deep learning to assist and enhance clinical POAG diagnosis. GlaucomaNet is publicly available on . com/ bionl plab/ Glauc omaNet.
Loading Preview
Sorry, preview is currently unavailable. You can download the paper by clicking the button above.