id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
cord-186831-724br56j	Chen, Xiaocong	Momentum Contrastive Learning for Few-Shot COVID-19 Diagnosis from Chest CT Images	2020-06-16		.txt	text/plain	4248	284	60	To this end, we propose a new deep learning algorithm for the automated diagnosis of COVID-19, which only requires a few samples for training. To this end, we propose a new deep learning algorithm for the automated diagnosis of COVID-19, which only requires a few samples for training. Specifically, we use contrastive learning to train an encoder which can capture expressive feature representations on large and publicly available lung datasets and adopt the prototypical network for classification. Specifically, we use contrastive learning to train an encoder which can capture expressive feature representations on large and publicly available lung datasets and adopt the prototypical network for classification. Finally, we utilized two public lung datasets to pre-train an embedding network and employ the prototypical network (Snell et al., 2017) to conduct the few-shot classification, which learns a metric space where the classification can be performed by measuring the distances to the derived prototypical representation of each class.	./cache/cord-186831-724br56j.txt	./txt/cord-186831-724br56j.txt