Skip to main content
. 2021 Jan 9;16(2):197–206. doi: 10.1007/s11548-020-02305-w

Table 6.

Comparison of proposed method with recent state-of-the-art methods for COVID-19 detection using CXR images

Study Method Dataset Acc (%)
Wang et al. [4] COVID-Net Training data Testing data: 93.3
7966 Normal 100 Normal
5438 Pneumonia 100 Pneumonia
258 COVID-19 100 COVID-19
Ozturk et al. [6] DarkCovidNet 500 Normal 87.02
500 Pneumonia
127 COVID-19
Haghanifar et al. [12] UNet+DenseNet Training data Testing data: 87.21
3000 Normal 724 Normal
3400 Pneumonia 672 Pneumonia
400 COVID-19 144 COVID-19
Siddhartha and COVIDLite 668 Normal 96.43
Santra [10] 619 Viral Pneumonia
536 COVID-19
Apostolopoulos and Mpesiana [13] VGG19 Testing data 1 Testing data 2: 93.48 & 94.72
504 Normal 504 Normal
700 Bacterial 714 Viral&
Pneumonia Bacterial Pneumonia
224 COVID-19 224 COVID-19
Proposed Method Fus-ResNet50 Testing data 1 Testing data 2 95.57&94.44
2567 Normal 6284 Normal
2567 Pneumonia 3478 Pneumonia
2567 COVID-19 756 COVID-19