DAFLNet: Dual Asymmetric Feature Learning Network for COVID-19 Disease Diagnosis in X-Rays

Algorithm 2

Pseudocode of DAFLNet algorithm.

Phase I: Preprocessing

Step 1 Input: Original Image . See Eq. (1).

Step 2 Dataset resizing. See Eq. (2).

Step 3 is divided into training set A, validation set B and test set C. See Eq. (6).

Step 4 MDA(A): noise injection, rotation, gamma correction and mirror to training set A.

Phase II: DAFFM

Step 5 Input of raw pre-trained EfficientNetV2 and DenseNet models.

Step 6 Obtain MBConv and Fused-MBConv Networks from .

Step 7 For

 Step 8 Add CA and SA to (CBAM).

 Step 9 Obtain residual Networks from .

 Step 10 . See Eq. (3).

 Step 11 Generate DAFLNET-1 and DAFLNET-2 based on the fusion model concat parameter .

 End

Step 12 Generate DAFFM.

Phase III: WDFM

Step 13 Obtain the and fusion features from .

Step 14 Obtain the .

Step 15 Obtain the weights of according to

Step 16 Concatenate to obtain the . See Eq. (5).

Step 17 Test confusion matrix, calculate indicators.

Step 18 Output: The DAFLNet model and its performances.