G-Eyenet: A Convolutional Autoencoding Classifier Framework for the Detection of Glaucoma from Retinal Fundus Images

Pal, Abhishek; Moorthy, Manav Rajiv; Shahina, A.

doi:10.1109/icip.2018.8451029

Cited by 44 publications

(43 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, the learned models might exhibit a weak generalization ability. To partially bypass this issue, some authors have proposed to train their methods using combinations of different data sets (Cerentinia et al, 2018;Pal et al, 2018). Table 1, all existing data sets with OD/OC annotations contain manually assigned labels obtained from the CFP, without considering depth information and performed by a single reader.…”

Section: Evaluation Protocolsmentioning

confidence: 99%

REFUGE Challenge: A unified framework for evaluating automated methods for glaucoma assessment from fundus photographs

Orlando

Breda

et al. 2020

Medical Image Analysis

536

248

View full text Add to dashboard Cite

Glaucoma is one of the leading causes of irreversible but preventable blindness in working age populations.Color fundus photography (CFP) is the most cost-effective imaging modality to screen for retinal disorders.However, its application to glaucoma has been limited to the computation of a few related biomarkers such as the vertical cup-to-disc ratio. Deep learning approaches, although widely applied for medical image analysis, have not been extensively used for glaucoma assessment due to the limited size of the available data sets. Furthermore, the lack of a standardize benchmark strategy makes difficult to compare existing methods in a uniform way. In order to overcome these issues we set up the Retinal Fundus Glaucoma Challenge, REFUGE (https://refuge.grand-challenge.org), held in conjunction with MICCAI 2018. The challenge consisted * Corresponding authors: Yanwu Xu (ywxu@ieee.org) and Xiulan Zhang (

show abstract

Section: Evaluation Protocolsmentioning

confidence: 99%

REFUGE Challenge: A unified framework for evaluating automated methods for glaucoma assessment from fundus photographs

Orlando

Breda

et al. 2020

Medical Image Analysis

536

248

View full text Add to dashboard Cite

show abstract

“…Table III and Table IV discussed performance of the proposed GlaucoNet+ model with different classification setups (Table III) and configuration (Table IV); however to further examine efficacy of our proposed model in comparison to the other existing glaucoma detection and classification methods, we have performed analysis based on secondary resources (reviewing existing methods or allied papers). [49] 84.38 --- [54] 95.50 --- [60] --- [64] 88.00 --- [35] --98.60 - [36] 99.20 -86.00 - [65] ---- [47] 80.00 -95.00 - [48] 97.00 --- [57] 98 --- [39] 94.10 -91.80 - [75] 90.00 --- [40] --92.00 - [33] 100.00 -94.00 - [59] ---- [66] 89.6 (NB) 97.6(AN N) --- [67] 92.00 --- [68] ---- [69] 72.38 --- [70] 79.00 -87.00 - [71] 83.10 --- [62] 93.00 --- [72] 96.67 -100.00 - [73] 91.00 --- [74] 92.00 --- [4] 0.8478 - [55] 88. Observing the results, it can be found that the proposed GlaucoNet+ model with Hybrid feature extraction and SVM (polynomial) with 10-fold cross validation outperforms major existing approaches.…”

Section: Resultsmentioning

confidence: 99%

“…To achieve it, they recommended deep learning methods with ability to exploit image-relevant information to make Glaucoma classification. In [59] multi-model network known as G-EyeNet was developed that comprised a deep convolutional auto-encoder (CAE) to perform Glaucoma detection. Authors [60] applied CNN for automatic Glaucoma detection and found it suitable for target detection purposes [61] [62].…”

Section: Related Workmentioning

confidence: 99%

Hybrid CNN assisted Computer Aided Diagnosis System for Glaucoma Detection and Classification: GlaucoNet+

2019

IJITEE

View full text Add to dashboard Cite

The exponential rise in technologies has revitalized academia-industries to achieve more efficient computer aided diagnosis systems. It becomes inevitable especially for Glaucoma detection which has been increasing with vast pace globally. Most of the existing approaches employs morphological features like optical disk and optical cup information, optical cup to disk ratio etc; however enabling optimal detection of such traits has always been challenge for researchers. On the other hand, in the last few years deep learning methods have gained widespread attention due to its ability to exploit fine grained features of images to make optimal classification decision. However, reliance of such methods predominantly depends on the presence of deep features demanding suitable feature extraction method. To achieve it major existing approaches extracts full-image features that with high dimensional kernel generates gigantically huge features, making classification computationally overburdened. Therefore, retaining optimal balance between deep features and computational overhead is of utmost significance for glaucoma detection and classification. With this motive, in this paper a novel hybrid deep learning model has been developed for Glaucoma detection and classification. The proposed Hybrid CNN model embodies Stacked Auto-Encoder (SAE) with transferable learning model AlexNet that extracts high dimensional features to make further two-class classification. To achieve computational efficiency, In addition to the classical ReLu and dropout (50%), we used Principle Component Analysis (PCA) and Linear Discriminant Analysis (LDA) algorithms. We applied 10-fold cross validation assisted Support Vector Machine classifier to perform two-class classification; Glaucomatous and Normal fundus images. Simulation results affirmed that the proposed Hybrid deep learning model with LDA feature selection and SVM-Poly classification achieves the maximum accuracy of 98.8%, precision 97.5%, recall 97.5% and F-Measure of 97.8%.

show abstract

“…In [35] Abhishek Pal et al, in 2018, proposed G-EYENET named autoencoding system. It consists of two model system frame.…”

Section: Comparison Of Various Automated Glaucoma Detection Technmentioning

confidence: 99%

Recent Automated Glaucoma Detection Techniques using Color Fundus Images

2019

IJITEE

View full text Add to dashboard Cite

One of the areas in which C-DAC, Mohali is actively engaged, is development of AI powered fundus imaging system providing insight into several severe eye diseases. Glaucoma, one of the most hazardous ocular disease, continues to affect and burden a large section of our population. Neuropathy of optic nerve cells is the prime cause of glaucoma and is the second leading cause of blindness worldwide. It doesn’t manifest itself and is often termed as the silent thief of eye sight. The damage caused by glaucoma is irreversible. Therefore, it is imperative to detect glaucoma at an early stage. The medical literature related to glaucoma indicates that glaucoma detection is a complex process and depends on combination of several parameters. The conventional methods of hand-crafted feature extraction are tedious, time consuming and require human intervention. Even though many such systems have recently shown promising results, but these systems require extensive feature engineering and have limited representation power owing to varied morphology of the optic nerve head. Most of the proposed systems have targeted the parameter cup to disc ratio (CDR) for detection of glaucoma, but that may not be the best approach for building efficient, robust and accurate automated system for glaucoma diagnosis. This paper advocates the use of hybrid approach of manual feature crafting with deep learning. It holds promise of improving the accuracy of glaucoma diagnosis through the automated techniques. It is further proposed that if diagnosis based on CDR remains inconclusive other methods of diagnosis should be adopted to come to a certain conclusion.

show abstract

G-Eyenet: A Convolutional Autoencoding Classifier Framework for the Detection of Glaucoma from Retinal Fundus Images

Cited by 44 publications

References 12 publications

REFUGE Challenge: A unified framework for evaluating automated methods for glaucoma assessment from fundus photographs

REFUGE Challenge: A unified framework for evaluating automated methods for glaucoma assessment from fundus photographs

Hybrid CNN assisted Computer Aided Diagnosis System for Glaucoma Detection and Classification: GlaucoNet+

Recent Automated Glaucoma Detection Techniques using Color Fundus Images

Contact Info

Product

Resources

About