CataractNet: An Automated Cataract Detection System Using Deep Learning for Fundus Images

Junayed, Masum Shah; Islam, Md Baharul; Sadeghzadeh, Arezoo; Rahman, Saimunur

doi:10.1109/access.2021.3112938

Cited by 81 publications

(37 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Junayed et al [23] presented CataractNet, a novel convolutional neural network architecture characterized by a reduced number of layers, lower training parameters, and compact kernel sizes. This design strategy aimed to decrease computational time and expense while retaining high accuracy in cataract classification, achieving an accuracy rate of 99.13%.…”

Section: Cataract Detection Using Deep Learning Methodsmentioning

confidence: 99%

Enhancing Cataract Detection Precision: A Deep Learning Approach

Yadav,

Yadav

2023

View full text Add to dashboard Cite

Cataracts, characterized by the opacification of the eye lens leading to visual deterioration, pose a significant global health issue. Timely and accurate detection of cataracts is pivotal for halting disease progression and augmenting the patients' quality of life. However, conventional diagnostic approaches for cataract detection and grading rely heavily on the expertise of ophthalmologists, a solution that can be unduly costly and inaccessible for certain population segments seeking early intervention. Addressing this challenge, the present study introduces a computer-assisted diagnostic strategy for the detection and grading of cataracts, drawing on fundus retinal images. The proposed approach capitalizes on a deep convolutional neural network to extract features from fundus images, which are subsequently evaluated via three distinct classification algorithms: Support Vector Machine, Naive Bayes, and Decision Tree. The resultant categorization stratifies the images into four severity levels: mild, moderate, normal, and severe. Further enhancing the classifier's prediction accuracy, an Ensemble (ES) learning mechanism via a Majority Voting Scheme (MVS) process is incorporated into the study. A total of 1600 fundus images, sourced from various open-access databases and classified into four categories by an expert ophthalmologist, were utilized for the study. The proposed methodology demonstrated a commendable accuracy rate of 97.34% in the four-stage cataract classification and grading, outperforming existing methodologies. This research advances the field by introducing a reliable, cost-effective, and accessible solution for early cataract detection, contributing significantly to global health improvements.

show abstract

Section: Cataract Detection Using Deep Learning Methodsmentioning

confidence: 99%

Enhancing Cataract Detection Precision: A Deep Learning Approach

Yadav,

Yadav

2023

View full text Add to dashboard Cite

show abstract

“…Contrary to conventional computer vision methods, CNN-based models extract more and deeper features which enhance the classification accuracy and enable the system to deal with more classification types [11]. Hence, they are the focus of interest among the researchers not only for acne detection and classification but also in every field of medical image analysis [23,8,26,22,38,28,17,33,18,19,20].…”

Section: Deep Learning-based Methodsmentioning

confidence: 99%

Acne Vulgaris Detection and Classification: A Dual Integrated Deep CNN Model

Islam,

Junayed,

Sadeghzadeh

et al. 2023

IJCAI

View full text Add to dashboard Cite

Recognizing acne disease and evaluating its type is vital for the efficacy of the medical treatment. This report collects a dataset of 420 images and then labels them into seven different classes by a well-experienced dermatologist. After a pre-processing step, including local and global contrast enhancement and noise removal by a smoothing filter, the dataset size is enhanced using augmentation. The images of the dataset and the augmented ones are all fed into a novel integrated dual deep convolutional neural network (CNN) model to recognize acne disease and its type by classifying it into seven groups. First, two CNN-based units are designed to extract deep feature maps, later combined in a feature aggregation module. The aggregated features provide rich input information and classify the acne by a softmax. The proposed architecture's optimizer, loss function, and activation functions are all tuned so that both CNN units are trained with minimum kernel size and fewer training parameters. Thus, the computational cost is minimized. Compared with three machine learning-based classifiers and five pre-trained models, our model achieves competitive state-of-the-art performance with an accuracy of 97.53% on the developed dataset.Povzetek: Razviti sistem uporablja CNN model za prepoznavanje in klasifikacijo aken (acne vulgaris) z natančnostjo 97.53% na naboru 420 slik.

show abstract

“…A novel deep learning model, named as CataractNet, was proposed by [21] to identify cataract in a dataset composed of 1,130 fundus images (augmented to 4,746 images). The model obtained accuracies higher than 98%, being the best literature result.…”

Section: A Cataractmentioning

confidence: 99%

Deep Transfer Learning Strategy to Diagnose Eye-Related Conditions and Diseases: An Approach Based on Low-Quality Fundus Images

2023

View full text Add to dashboard Cite

Data from the World Health Organization indicate that billion cases of visual impairment could be avoided, mainly with regular examinations. However, the absence of specialists in basic health units has resulted in a lack of accurate diagnosis of systemic or asymptomatic eye diseases, increasing the cases of blindness. In this context, the present paper proposes an ensemble of convolutional neural networks, which were submitted to a transfer learning process by using 38,727 high-quality fundus images. Next, the ensemble was tested with 13,000 low-quality fundus images acquired by low-cost equipment. Thus, the proposed approach contributes to advance the state-of-the-art with a novel deep transfer learning strategy, which is more suitable and feasible to be applied by emerging and under-developing countries. From lowquality images, the proposed approach was able to reach accuracies of 87.4%, 90.8%, 87.5%, 79.1% to classify cataract, diabetic retinopathy, excavation and blood vessels, respectively.INDEX TERMS Convolutional neural network; deep learning; eye-related conditions; fundus images; transfer learning.

show abstract

CataractNet: An Automated Cataract Detection System Using Deep Learning for Fundus Images

Cited by 81 publications

References 46 publications

Enhancing Cataract Detection Precision: A Deep Learning Approach

Enhancing Cataract Detection Precision: A Deep Learning Approach

Acne Vulgaris Detection and Classification: A Dual Integrated Deep CNN Model

Deep Transfer Learning Strategy to Diagnose Eye-Related Conditions and Diseases: An Approach Based on Low-Quality Fundus Images

Contact Info

Product

Resources

About