This research article presents a hybrid deep learning CNN model for the diagnosis of COVID-19 using chest X-ray images. The model combines a heading model and a base model, utilizing pre-trained deep learning structures like VGG16 and VGG19. The feature dimensions from these models are reduced using different pooling layers (max and average), and dense layers with various activation functions are added to the heading part. A dropout layer is also included to prevent overfitting. The model's performance is evaluated using a COVID-19 radiology database and compared with existing transfer learning architectures such as VGG16, VGG19, EfficientNetB0, and ResNet50. Various classification techniques, including K-Nearest Neighbor (KNN), Naïve Bayes, Random Forest, Support Vector Machine (SVM), and Neural Network, are used for performance comparison. The hybrid deep learning model, particularly with average pooling layers and SVM-linear, achieves an accuracy of 92%, outperforming other models in terms of precision, recall, and F1 score. The proposed model can assist radiologists and physicians in reducing misdiagnosis rates and validating positive COVID-19 cases. The study also discusses the importance of early diagnosis of COVID-19 and the challenges posed by the pandemic, highlighting the need for efficient and accurate diagnostic tools.This research article presents a hybrid deep learning CNN model for the diagnosis of COVID-19 using chest X-ray images. The model combines a heading model and a base model, utilizing pre-trained deep learning structures like VGG16 and VGG19. The feature dimensions from these models are reduced using different pooling layers (max and average), and dense layers with various activation functions are added to the heading part. A dropout layer is also included to prevent overfitting. The model's performance is evaluated using a COVID-19 radiology database and compared with existing transfer learning architectures such as VGG16, VGG19, EfficientNetB0, and ResNet50. Various classification techniques, including K-Nearest Neighbor (KNN), Naïve Bayes, Random Forest, Support Vector Machine (SVM), and Neural Network, are used for performance comparison. The hybrid deep learning model, particularly with average pooling layers and SVM-linear, achieves an accuracy of 92%, outperforming other models in terms of precision, recall, and F1 score. The proposed model can assist radiologists and physicians in reducing misdiagnosis rates and validating positive COVID-19 cases. The study also discusses the importance of early diagnosis of COVID-19 and the challenges posed by the pandemic, highlighting the need for efficient and accurate diagnostic tools.