id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
cord-256008-lwki1rzc	Sekeroglu, Boran	Detection of COVID-19 from Chest X-Ray Images Using Convolutional Neural Networks	2020-09-18		.txt	text/plain	6949	348	46	When the images fed ConvNets directly (Experiments 11-17), we observed that the increment of the convolutional layer number of ConvNets reduces the scores obtained by the neural network up to 4%, similar to COVID-19/Normal results. Similar results were obtained in the experiments, and nB produced the highest mean ROC AUC, mean sensitivity, and mean accuracy scores (88.92, 80.00, and 96.96%, respectively) for statistical measurement experiments of COVID-19/Pneumonia classification. Inception-V3 produced higher results than other pre-trained networks; however, the highest mean ROC AUC score in transfer learning experiments was obtained by DenseNet121 (96.48%). In COVID-19/Normal classification, the highest mean specificity (when the 100.0% scores of pre-trained networks are not considered because of not learning another class) and the highest mean accuracy results were obtained in Exp.14 (99.78 and 99.11%, respectively), which consisted of the deepest architecture in ConvNet experiments ( Table 4 ).	./cache/cord-256008-lwki1rzc.txt	./txt/cord-256008-lwki1rzc.txt