Journal: Scientific Reports

Article Title: A comparative evaluation of deep learning approaches for ophthalmology

doi: 10.1038/s41598-024-72752-x

Figure Lengend Snippet: Accuracy of each architecture for each dataset for multiclass problem (> 90% dark, > 80% medium otherwise light).

Article Snippet: The comparison showed that there are few disadvantages to using grading over multiclass classification, especially since the accuracy is similar (85% for EfficientNet multiclass Eyepacs versus 88% for grading).

Techniques:

Journal: Scientific Reports

Article Title: A comparative evaluation of deep learning approaches for ophthalmology

doi: 10.1038/s41598-024-72752-x

Figure Lengend Snippet: Original ACRIMA glaucoma image and the heatmap generated from EfficientNet.

Article Snippet: The comparison showed that there are few disadvantages to using grading over multiclass classification, especially since the accuracy is similar (85% for EfficientNet multiclass Eyepacs versus 88% for grading).

Techniques: Generated

Journal: Scientific Reports

Article Title: A comparative evaluation of deep learning approaches for ophthalmology

doi: 10.1038/s41598-024-72752-x

Figure Lengend Snippet: List of architectures and accuracy (in %) of each dataset for multiclass problem.

Article Snippet: The comparison showed that there are few disadvantages to using grading over multiclass classification, especially since the accuracy is similar (85% for EfficientNet multiclass Eyepacs versus 88% for grading).

Techniques:

Journal: Scientific Reports

Article Title: A comparative evaluation of deep learning approaches for ophthalmology

doi: 10.1038/s41598-024-72752-x

Figure Lengend Snippet: Performance metrics of grading classifiers.

Article Snippet: The comparison showed that there are few disadvantages to using grading over multiclass classification, especially since the accuracy is similar (85% for EfficientNet multiclass Eyepacs versus 88% for grading).

Techniques:

Journal: Scientific Reports

Article Title: A comparative evaluation of deep learning approaches for ophthalmology

doi: 10.1038/s41598-024-72752-x

Figure Lengend Snippet: Accuracy (in %) of CNN classifiers on different 2D OCT datasets.

Article Snippet: The comparison showed that there are few disadvantages to using grading over multiclass classification, especially since the accuracy is similar (85% for EfficientNet multiclass Eyepacs versus 88% for grading).

Techniques:

Journal: Multimedia Tools and Applications

Article Title: Recent developments on computer aided systems for diagnosis of diabetic retinopathy: a review

doi: 10.1007/s11042-022-13841-9

Figure Lengend Snippet: Methods to detect hemorrhages

Article Snippet: Supriya Mishra et al. [ ] , 2020 , Kaggle (APTOS) , DL DenseNet Architecture , Accuracy- 0.9611 Kappa Score- 0.8981 , Multiclass (5- Classes).

Techniques: Transformation Assay, Sampling, Modification

Journal: Multimedia Tools and Applications

Article Title: Recent developments on computer aided systems for diagnosis of diabetic retinopathy: a review

doi: 10.1007/s11042-022-13841-9

Figure Lengend Snippet: Different techniques for categorization of DR

Article Snippet: Supriya Mishra et al. [ ] , 2020 , Kaggle (APTOS) , DL DenseNet Architecture , Accuracy- 0.9611 Kappa Score- 0.8981 , Multiclass (5- Classes).

Techniques: Labeling, Biomarker Discovery, Transformation Assay

Journal: Multimedia Tools and Applications

Article Title: Recent developments on computer aided systems for diagnosis of diabetic retinopathy: a review

doi: 10.1007/s11042-022-13841-9

Figure Lengend Snippet: Different types of feature selection and fusion methods

Article Snippet: Supriya Mishra et al. [ ] , 2020 , Kaggle (APTOS) , DL DenseNet Architecture , Accuracy- 0.9611 Kappa Score- 0.8981 , Multiclass (5- Classes).

Techniques: Selection, Extraction

Journal: Multimedia Tools and Applications

Article Title: Recent developments on computer aided systems for diagnosis of diabetic retinopathy: a review

doi: 10.1007/s11042-022-13841-9

Figure Lengend Snippet: Adversarial attack for image classification

Article Snippet: Supriya Mishra et al. [ ] , 2020 , Kaggle (APTOS) , DL DenseNet Architecture , Accuracy- 0.9611 Kappa Score- 0.8981 , Multiclass (5- Classes).

Techniques: CRAfT Assay, Imaging