Detecting COVID-19 in Chest X-Ray Images via MCFF-Net

Abstract

COVID-19 is a respiratory disease caused by severe acute respiratory syndrome coronavirus (SARS-CoV-2). Due to the rapid spread of COVID-19 around the world, the number of COVID-19 cases continues to increase, and lots of countries are facing tremendous pressure on both public and medical resources. Although RT-PCR is the most widely used detection technology with COVID-19 detection, it still has some limitations, such as high cost, being time-consuming, and having low sensitivity. According to the characteristics of chest X-ray (CXR) images, we design the Parallel Channel Attention Feature Fusion Module (PCAF), as well as a new structure of convolutional neural network MCFF-Net proposed based on PCAF. In order to improve the recognition efficiency, the network adopts 3 classifiers: 1-FC, GAP-FC, and Conv1-GAP. The experimental results show that the overall accuracy of MCFF-Net66-Conv1-GAP model is 94.66% for 4-class classification. Simultaneously, the classification accuracy, precision, sensitivity, specificity, and F1-score of COVID-19 are 100%. MCFF-Net may not only assist clinicians in making appropriate decisions for COVID-19 diagnosis but also mitigate the lack of testing kits.

Role of Artificial Intelligence in COVID-19 Detection

“…It is also observed that the two-class models are no longer valid when other diseases with similar symptoms were presented [ 178 ]. It is noted from Table 2 , Table 3 , Table 4 and Table 5 that few studies have performed four-class (normal vs. COVID-19 vs. viral pneumonia (VP) vs. bacterial pneumonia (BP)) classification [ 114 , 118 , 138 , 154 , 161 , 179 , 189 , 194 , 264 ]. They have obtained the Cvd.Acc (avg.)…”

“…Further, we also analyzed the systems which can categorize three or more classes. It is observed from Table 2 that ResNet50 with DWT and GLCM [ 114 ], customized CNN [ 118 , 154 , 179 , 189 ], GoogLeNet [ 138 ], InceptionNet [ 141 ], AlexNet [ 160 ], a combination of DenseNet103 and ResNet18 [ 148 ], an ensemble of various models such as InceptionResNetV2, ResNet152V2, VGG16, and DenseNet201 [ 153 ], and a grouping of MobileNet and InceptionV3 [ 161 ] were effectively used for four-class classification using X-ray images. Further, the authors also used CNN models for five-class classification using X-ray images [ 129 , 168 , 177 ].…”

A COVID‐19 CXR image recognition method based on MSA‐DDCovidNet

“…To recognize the COVID‐19 CXR images, Wang et al. [ 15 ] designed a new network MCFF‐Net based on the Parallel Channel Attention Feature Fusion Module (PCAF). Wang et al.…”