Classification of Diabetic Retinopathy Severity in Fundus Images Using the Vision Transformer and Residual Attention

Gu, Zongyun; Li, Yan; Wang, Zijian; Kan, Junling; Jianhua, Shu; Wang, Qing

doi:10.1155/2023/1305583

Cited by 45 publications

(8 citation statements)

References 63 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some of these use cases are illustrated by the following research papers. For example, Manali Gupta et al [6] implemented and evaluated the performance of a scratch CNN method with VGG-16 for classification of brain tumor MRI images as cancerous or non-cancerous. Samir S. Yadav et al [7] evaluated convolutional neural network based architectures for classification of pneumonia presence on chest X-Ray images dataset.…”

Section: Existing Work Explainability Of Classification Models In The...mentioning

confidence: 99%

Explainable AI for Interpretation of Ovarian Tumor Classification Using Custom ResNet60

Guha,

Kodipalli,

Fernandes

et al. 2024

Preprint

View full text Add to dashboard Cite

Deep learning architectures like ResNet and Inception have produced accurate predictions for classifying benign and malignant tumors in the healthcare domain. This enables healthcare institutions to make data-driven decisions and potentially enable early detection of malignancy by employing computer-vision-based deep learning algorithms. These Convolutional Neural Network algorithms, in addition to requiring huge amounts of data, can identify higher and lower-level features that are significant while classifying tumors into benign or malignant. However, these algorithms have limitations concerning explainability and identifying the exact features that are of importance and contributing to the final classification of tumors as benign or malignant. In this paper, we implement several explainable AI techniques, namely, LIME, Saliency Map, Occlusion Analysis, Grad-CAM, SHAP, and Smooth Grad to interpret the results of a custom ResNet60 classifier in the classification of ovarian tumors as benign or malignant. The ResNet60 model attained an accuracy of 97.50% for the test dataset.

show abstract

Section: Existing Work Explainability Of Classification Models In The...mentioning

confidence: 99%

Explainable AI for Interpretation of Ovarian Tumor Classification Using Custom ResNet60

Guha,

Kodipalli,

Fernandes

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…The GPB's residual attention was well-suited in identifying the areas of space inhabited by various types of objects. Extensive studies using DDR datasets proved that the technique is superior, and when compared to the gold standard, it performed comparably [15]. 2022) use a nested U-Net to segment red lesions and subsequently filter out false positives using a sub-image classification method.…”

Section: Related Workmentioning

confidence: 99%

An Efficient Feature Selection and Extraction using Metaheuristic Technique for Diabetic Retinopathy

Anamika Raj

2024

jes

View full text Add to dashboard Cite

Diabetic retinopathy (DR) is associated with diabetes, which causes harm to the retina as a result of persistent elevated blood sugar levels, resulting in symptoms such as impaired vision. Regular eye examinations are crucial for the timely identification of any issues. The process of selecting characteristics in DR involves optimizing important features using a combination of Hybridized Weight-Optimized Particle Swarm Optimization and Whale Optimization Algorithm (HWOPSO-WOA). This leads to an enhancement in the accuracy of detection, which is crucial for appropriate diagnosis and therapy. Pre-processing encompasses the utilization of linear filters such as Sobel and Prewitt for image manipulation, mean filters for the reduction of noise, and Gaussian filters for the purpose of smoothing. Feature selection employs an objective function that considers significant performance measure, which utilizes binary encoding and conducts fitness evaluation. The methodology is utilized for the analysis of DR utilizing the Indian Diabetic Retinopathy Image Dataset. The results comprise convergence graphs, and a comparative analysis, which emphasize the greater accuracy and efficiency of the Proposed technique. Visual representations, such as fundus images with selected features, highlight the importance of the chosen features in detecting DR. The proposed HWOPSO-WOA achieves highest accuracy of 97.3% and minimal processing time 98.45seconds that outperforms the state-of-art techniques.

show abstract

“…Classification accuracies of 96.6%, 96.6%, 95.6%, and 96.6% were achieved for healthy images, stage 1, stage 2, and stage 3, respectively. Gu et al [ 11 ] proposed a method to detect all five DR stages, no DR, mild, moderate, severe, and proliferative DR. They employed a feature extraction block (FEB) and a grading prediction block (GPB) and used the DDR and the IDRiD datasets.…”

Section: Literature Reviewmentioning

confidence: 99%

A Computer-Aided Diagnostic System to Identify Diabetic Retinopathy, Utilizing a Modified Compact Convolutional Transformer and Low-Resolution Images to Reduce Computation Time

et al. 2023

View full text Add to dashboard Cite

Diabetic retinopathy (DR) is the foremost cause of blindness in people with diabetes worldwide, and early diagnosis is essential for effective treatment. Unfortunately, the present DR screening method requires the skill of ophthalmologists and is time-consuming. In this study, we present an automated system for DR severity classification employing the fine-tuned Compact Convolutional Transformer (CCT) model to overcome these issues. We assembled five datasets to generate a more extensive dataset containing 53,185 raw images. Various image pre-processing techniques and 12 types of augmentation procedures were applied to improve image quality and create a massive dataset. A new DR-CCTNet model is proposed. It is a modification of the original CCT model to address training time concerns and work with a large amount of data. Our proposed model delivers excellent accuracy even with low-pixel images and still has strong performance with fewer images, indicating that the model is robust. We compare our model’s performance with transfer learning models such as VGG19, VGG16, MobileNetV2, and ResNet50. The test accuracy of the VGG19, ResNet50, VGG16, and MobileNetV2 were, respectively, 72.88%, 76.67%, 73.22%, and 71.98%. Our proposed DR-CCTNet model to classify DR outperformed all of these with a 90.17% test accuracy. This approach provides a novel and efficient method for the detection of DR, which may lower the burden on ophthalmologists and expedite treatment for patients.

show abstract

Classification of Diabetic Retinopathy Severity in Fundus Images Using the Vision Transformer and Residual Attention

Cited by 45 publications

References 63 publications

Explainable AI for Interpretation of Ovarian Tumor Classification Using Custom ResNet60

Explainable AI for Interpretation of Ovarian Tumor Classification Using Custom ResNet60

An Efficient Feature Selection and Extraction using Metaheuristic Technique for Diabetic Retinopathy

A Computer-Aided Diagnostic System to Identify Diabetic Retinopathy, Utilizing a Modified Compact Convolutional Transformer and Low-Resolution Images to Reduce Computation Time

Contact Info

Product

Resources

About