Ith results shown in Table 1. The increase in accuracy is exceptionally critical when it involves the diagnosis of a significant health-related condition for instance COVID-19. The highest reported accuracy was again accomplished applying VGG16 with an average value of 94.26 , precision of 99 , recall of 99.18 , and F1-Measure of 99.09 . The lowest accuracy is again reported applying the Hexazinone MedChemExpress GoogleNet with an average value of 90.38 .Table four. Experimental results of diverse models with nonfreeze weights and augmented normalized information.Accuracy VGG19 VGG16 DenseNet AlexNet GoogleNet 94.07 94.26 92.23 91.28 90.Precision 99.12 99.00 98.71 97.68 97.Recall 99.56 99.18 98.89 98.32 96.F1 Measure 99.34 99.09 98.80 98.00 97.Returning towards the original non-normalized information just after applying the enhancement algorithm on it and utilizing the non-freeze weights, the experimental final results employing the optimized DL models are shown in Table five. This scenario offers the most effective benefits as compared to that on the experiment shown in Table 1. As is usually observed, the enhancement of pictures increased accuracy significantly compared with those reported in Table 1, using the highest accuracy achieved by VGG19 with an average value of 94.92 , precision of 99.37 , recall of 99.28 , and F1-measure of 99.33 . The lowest accuracy is reported using the GoogleNet, with an typical worth of 89.two .Table 5. Experimental outcomes of distinct models with nonfreeze weights and enhanced non-normalized data.Accuracy VGG19 VGG16 DenseNet AlexNet GoogleNet 94.92 94.26 91.94 91.87 89.Precision 99.37 99.00 97.67 96.52 97.Recall 99.28 99.18 99.09 98.84 97.F1 Measure 99.33 99.09 98.37 97.67 97.Diagnostics 2021, 11,13 ofThe experiments are repeated working with the enhanced normalized data with nonfreeze weights applying the optimized DL models and shown in Table six. The results are much better as in comparison with the results obtained in Table 2. Once more, we observe a rise in accuracy, with the highest reported accuracy making use of the VGG16 with an average worth of 94.98 , precision of 100 , recall of 97.63 , and F1-measure of 98.eight . The lowest accuracy is reported once again for GoogleNet, with an average value of 84.76 . Enhancement of data improved the accuracy for both normalized and non-normalized information.Table 6. Experimental outcomes of various models with nonfreeze weights and enhanced normalized information.Accuracy VGG19 VGG16 DenseNet AlexNet GoogleNet 93.68 94.98 87.64 89.50 84.Precision 98.76 one hundred 95.02 94.40 93.Recall 95.98 97.63 94.63 95.63 89.F1 Measure 97.35 98.80 94.82 95.01 91.Now combining the augmented enhanced normalized dataset along with nonfreeze weights, the contribution of this function becomes evident because the accuracy continues to DBCO-Maleimide Biological Activity enhance with all the highest typical accuracy of 95.63 achieved employing the VGG16 in conjunction with the precision of 99.18 , recall of 98.78 , and F1-Measure of 98.98 , as shown in Table 7. The lowest accuracy achievement continues to become for the GoogleNet with an typical value of 88.43 . These results show that using a sizable dataset, an acceptable larger degree of accuracy is accomplished applying the optimized DL models. Using a sizable dataset, they are a few of the highest accuracies reported when in comparison with these offered inside the extant literature.Table 7. Experimental results of distinct models with nonfreeze weights and enhanced normalized augmented data.Accuracy VGG19 VGG16 DenseNet AlexNet GoogleNet 94.61 95.63 91.47 92.96 88.Precision 99.09 99.18 96.38 96.37 94.Recall 98.10 98.78 96.18 96.86 92.F1 Measure 98.60 98.98 96.28 96.61 93.The.