COVID-19 Diagnosis Using an Enhanced Inception-ResNetV2 Deep Learning Model in CXR Images

Table 3

Inception-ResNetV2 model architecture.

Layer	Patch size	Input size
Conv	3 × 3	224 × 224 × 3
Conv	3 × 3	111 × 111 × 32
Filter contact	3 × 3 pool + 3 × 3 conv	109 × 109 × 64
Filter contact	1 × 1 conv, 3 × 3 conv+ 1 × 1 conv, 7 × 1 conv, 1 × 7 conv, 3 × 3 conv	54 × 54 × 160
Filter contact	3 × 3 conv + max pool	52 × 52 × 128
Inception-ResNet-A × 10	—	26 × 26 × 256
Reduction-A	—	26 × 26 × 256
Inception-ResNet-B × 20	—	13 × 13 × 768
Reduction-B	—	13 × 13 × 768
Inception-ResNet-C × 10	—	6 × 6 × 1534
Average pooling	6 × 6	6 × 6 × 1534
Dropout	Keep = 0.5	1 × 1 × 1534
Fc	1534 × 1000	1534
Fc	1000 × 3	1000
Softmax	Classifier (3 classes)	500