Deep learning for identifying and classifying retinal diseases

Berrimi, Mohamed; Moussaoui, Abdelouaheb

doi:10.1109/iccis49240.2020.9257674

Cited by 13 publications

(8 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, it is remarked that pre-trained CNN architectures cannot provide the expected high performance in the classification of retinal diseases, and they show lower performances or close performances to each other than the CNN architectures designed in terms of diagnosing the retinal diseases. Among the studies compared, there are methods whose calculation cost is close to or lower than the proposed method (Alqudah, 2020;Apon et al, 2021;Berrimi & Moussaoui, 2020;Bhowmik et al, 2019;Najeeb et al, 2019;Tayal et al, 2021). However, surveying the results attained, it may be spotted that the performance loss is high.…”

Section: Resultsmentioning

confidence: 99%

“…At the end of the study, they achieved 95.66% accuracy and 95% sensitivity. Berrimi and Moussaoui (2020) introduced two new CNN architectures to detect retinal diseases from OCT scans. They used the OCT dataset containing four classes published by Zhang Lab at the University of California at San Diego (UCSD) (Kermany et.…”

Section: Literature Reviewmentioning

confidence: 99%

“…As a result of the trainings, Inception V3 architecture exhibited the best result with 99.27% accuracy. When the evaluation results were ranked from the best to the worst, the second CNN architecture designed by Berrimi and Moussaoui (2020) came after the Inception V3 architecture with an accuracy rate of 98.65%. Bhadra and Kar (2020) proposed a deep multi-layered CNN for retinal disease detection and classification.…”

Section: Literature Reviewmentioning

confidence: 99%

“…Because one of the disadvantages of deep learning models is that they need a lot of data for training. For this reason, architectures such as ImageNet, which are trained with big data before and achieve very successful results, are mostly used in the studies (Berrimi & Moussaoui, 2020;Bhowmik et al, 2019;Kermany et al 2018;Ye et al, 2019). These pre-trained models have learned how to analyze the image, and it is possible to retrain and exploit the model with less data for our problems by using the feature maps of these models.…”

Section: Classificationmentioning

confidence: 99%

See 3 more Smart Citations

Detection of retinal diseases from ophthalmological images based on convolutional neural network architecture

2022

View full text Add to dashboard Cite

The retina is an eye layer that incorporates light- and color-sensitive cells as well as nerve fibers. It collects light and distributes it to the brain for image processing through the use of the optic nerve. Diseases that end up causing vision loss and blindness are generated by retinal ailments. As a result, it is imperative to diagnose and treat certain disorders as early as possible. Optical coherence tomography (OCT), an angiography imaging technique, is operated to help diagnose retinal disorders. Deep learning approaches, which are extensively utilized, have now become a convenient way for diagnosing retinal illnesses through OCT images as a result of their effective outcomes in interpreting medical images. To diagnose retinal disorders utilizing OCT scans, this investigation developed a hybrid methodology based on image pre-processing and convolutional neural networks (CNNs) (a deep learning method). Image pre-processing techniques including background filling, resizing, noise reduction, and highlighting are exercised at the pre-processing stage. The segmentation process provides a new CNN architecture with five convolution layers that does have a low computational cost. Compared to other publications using the same data set, the proposed method seems to have a success rate of 99.48 percent in the detection of retinal disorders, closing a significant gap in the literature. The proposed approach has the advantage of maintaining low computing costs in comparison to other studies in the literature. When the conclusions are regarded, it is noticed that the suggested method might be exerted as a decision support system to assist physicians in the clinical context during the diagnosis of retinal disorders.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Literature Reviewmentioning

confidence: 99%

Section: Literature Reviewmentioning

confidence: 99%

Section: Classificationmentioning

confidence: 99%

See 2 more Smart Citations

Detection of retinal diseases from ophthalmological images based on convolutional neural network architecture

2022

View full text Add to dashboard Cite

show abstract

“…According to studies, the number of people living with macular degeneration is expected to reach 288 million worldwide by 2040 [2]. Soft drusen, which is a deposit of granular or amorphous material, is considered a precursor to AMD [3].…”

Section: __________________________mentioning

confidence: 99%

A Deep Learning-Based Technique for Diagnosing Retinal Disease by Using Optical Coherence Tomography (OCT) Images

Şahin

2022

Turkish Journal of Science and Technology

View full text Add to dashboard Cite

The retina is a network layer that covers the back wall of the human eye and is made up of visual cells. The retina layer is the most complex and sensitive part of the eye, and disorders that affect it have a big impact on people's lives. The Optical Coherence Tomography (OCT) imaging technology can be used to diagnose diseases that arise because of pathological alterations in the retina. The importance of early diagnosis in the management of these illnesses cannot be overstated. In this article, a approach based on convolutional neural networks (CNN), a deep learning method, is presented for the detecting of retinal disorders from OCT images, based on the literature review. For disease detection and categorization, a CNN architecture is developed. The proposed method has been found to have a 94 percent accuracy rate in the identification of retinal disorders. In a deep learning application for classification, the proposed CNN network model is compared to the MobileNet50 network model available in the literature, and the results are provided. The evaluation parameter values for models trained using the 5-fold cross-validation approach for each type of disease in the retinal OCT images dataset are also included in the publication. The proposed method can clearly be utilized as a decision-making tool to assist clinicians in diagnosing retinal illnesses in a clinical context, based on its effectiveness thus far.

show abstract

A Hybrid Images Deep Trained Feature Extraction and Ensemble Learning Models for Classification of Multi Disease in Fundus Images

Verma,

Kansal,

Popli

et al. 2024

Communications in Computer and Information Science

View full text Add to dashboard Cite

Retinal disorders, including diabetic retinopathy and macular degeneration due to aging, can lead to preventable blindness in diabetics. Vision loss caused by diseases that affect the retinal fundus cannot be reversed if not diagnosed and treated on time. This paper employs deep-learned feature extraction with ensemble learning models to improve the multi-disease classification of fundus images. This research presents a novel approach to the multi-classification of fundus images, utilizing deep-learned feature extraction techniques and ensemble learning to diagnose retinal disorders and diagnosing eye illnesses involving feature extraction, classification, and preprocessing of fundus images. The study involves analysis of deep learning and implementation of image processing. The ensemble learning classifiers have used retinal photos to increase the classification accuracy. The results demonstrate improved accuracy in diagnosing retinal disorders using DL feature extraction and ensemble learning models. The study achieved an overall accuracy of 87.2%, which is a significant improvement over the previous study. The deep learning models utilized in the study, including NASNetMobile, InceptionResNetV4, VGG16, and Xception, were effective in extracting relevant features from the Fundus images. The average F1-score for Extra Tree was 99%, while for Histogram Gradient Boosting and Random Forest, it was 98.8% and 98.4%, respectively. The results show that all three algorithms are suitable for the classification task. The combination of DenseNet feature extraction technique and RF, ET, and HG classifiers outperforms other techniques and classifiers. This indicates that using DenseNet for feature extraction can effectively enhance the performance of classifiers in the task of image classification.

show abstract

Deep learning for identifying and classifying retinal diseases

Cited by 13 publications

References 22 publications

Detection of retinal diseases from ophthalmological images based on convolutional neural network architecture

Detection of retinal diseases from ophthalmological images based on convolutional neural network architecture

A Deep Learning-Based Technique for Diagnosing Retinal Disease by Using Optical Coherence Tomography (OCT) Images

A Hybrid Images Deep Trained Feature Extraction and Ensemble Learning Models for Classification of Multi Disease in Fundus Images

Contact Info

Product

Resources

About