Multiclass Classification for Detection of COVID-19 Infection in Chest X-Rays Using CNN

Table 1

Illustration of similar studies according to the year, classification type, model, dataset, and performance.

Study	Year	Classification	Model	Dataset	Performance
Application of deep learning for fast detection of COVID-19 in X-Rays using nCOVnet [22]	2020	Binary	VGG-16 model with transfer learning	284 total images of COVID-19 and normal CXR	Accuracy: 88.10%; Sensitivity: 97.62%; Specificity: 78.57%
COVID-19 detection in chest X-Ray images using a new channel boosted CNN [23]	2022	Binary	Channel boosted split-transform-merge with region and edge-based operation	6,000 (COVID-healthy) 10,000 (COVID-viral infection) 15,000 (COVID-viral infection) CXR	Accuracy: 96.53%; F1-score: 95%
Chest X-Ray classification for the detection of COVID-19 using deep learning techniques [21]	2022	Multiclass	Efficient NetB1 with transfer learning	21,165 CXR of COVID-19, lungopacity, normal, and pneumonia	Accuracy: 96.13%; F1-score: 97.50%; Sensitivity: 9 6.50%
An efficient deep learning model to detect COVID-19 using chest X-Ray images [22]	2022	Multiclass	ResNet18 with transfer learning	10,040 CXR of COVID-19, pneumonia, and normal	Accuracy: 96.43%; Sensitivity: 93.68%; F1-score: 93%
MANet: A two-stage deep learning method for classification of COVID-19 from chest X-Ray images [23]	2021	Multiclass	ResNet50 with mask attention	6,792 CXR of COVID-19, viral pneumonia, bacterial pneumonia, tuberculosis, and normal	Accuracy: 96.03%; F1-score: 97%
X-Ray and CT-scan-based automated detection and classification of COVID-19 using convolutional neural networks (CNN) [24]	2021	Multiclass	CNN	6,077 CXR and CT of COVID-19, pneumonia, and normal	Accuracy: 98.28%; F1-score: 98.23%