id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
cord-207839-h8mcmqnc	Amran, Dor	Automated triage of COVID-19 from various lung abnormalities using chest CT features	2020-10-24		.txt	text/plain	2309	131	42	More specifically, we produce multiple descriptive features, including lung and infections statistics, texture, shape and location, to train a machine learning based classifier that distinguishes between COVID-19 and other lung abnormalities (including community acquired pneumonia). Figure 2 shows our system, which is composed of two main steps: an image processing pipeline and a feature analysis pipeline, which produces the final classification. As shown in Figure 2 , once the image pipeline outputs and feature extraction steps are performed for all chest CT scans, a feature-based dataset is used to train a machine learning classifier to differentiate between COVID-19 and other lung abnormalities. Rows 1 − 5: Classification performance (mean results) of different classifiers utilizing all depicted features. To assess the quantitative contribution of the groups we conducted an ablation study, removing in turn each of the feature groups from the classification process and measuring the effect on the final ensemble classifier performance.	./cache/cord-207839-h8mcmqnc.txt	./txt/cord-207839-h8mcmqnc.txt