|
[28]
|
U-net, Resnet-50-2D |
Classification, quantification and tracking: COVID-19 patients |
0.996 (AUC) 98.2% (sensitivity) 92.2% (specificity) |
AI based software |
|
[70]
|
ResNet-18 |
Feature extraction from image data |
73.1% (Accuracy) 67% (specificity) and 74% (sensitivity) |
A CNN based algorithm leveraging decision tree and SVM |
|
[42]
|
ResNet-50 |
Classification : COVID-19 and pneumonia |
0.96(AUC) |
A CNN based model: COVnet |
| [77]
|
Resnet50 |
Classification : COVID-19 |
95.38%(Accuracy), 95.52%(FPR), 91.41%(F1- score), 90.76%(kappa) |
A CNN based model |
|
[30]
|
VGG19, Mobile Net, Inception, Xception Inception ResNet v2 |
Classification: COVID-19, Model Evaluation |
97.82% (Accuracy) |
A proposal: best deep learning network |
|
[43]
|
VGG19, DenseNet121, ResNetV2, InceptionV3, InceptionResNetV2, Xception, and MobileNetV2 |
Classification:COVID-19, model evaluation |
F1-scores : normal :0.91 COVID-19 : 0.89 |
A proposal: best deep learning network |
|
[71]
|
InceptionV3 |
Classification:COVID-19 |
100% (Specificity) 100% (accuracy) 100% (PPV) 100%, (NPV) 100% (F-1 score) |
A proposed model, implementation and evaluation |
|
[31]
|
ResNet50, InceptionV3 and Inception-ResNetV2 |
Classification: COVID-19 |
98% (Accuracy) 96% (recall) and 100%(specificity) |
A proposal: best deep learning network |
| [38]
|
AH-Net |
Classification: COVID-19 |
Accuracy (90.8%), AUC(0.949) |
A proposal: best deeplearning network |
|
[40]
|
VGG-16, R-CNN |
Classification: COVID-19 |
Accuracy(97.36%),sensitivity (97.65%),precision (99.28%) |
A CNN based model |
