id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
cord-027303-20plzyqd	Krishnan, Gokul S.	Hybrid Text Feature Modeling for Disease Group Prediction Using Unstructured Physician Notes	2020-05-23		.txt	text/plain	3715	164	46	In this article, we propose a generic ICD9 disease group prediction CDSS built on unstructured physician notes modeled using hybrid word embeddings. In this article, a hybrid feature modeling approach that uses hybrid clinical word embeddings to generate quality features which are used to train and build a deep neural network model to predict ICD9 disease groups is presented. From our experiments, we observed a significant potential in developing prediction based CDSS using unstructured text reports directly, eliminating the dependency on the availability of structured patient data and EHRs. The proposed approach that involves a textual feature modeling and a neural network based prediction model was successful in capturing the rich and latent clinical information available in unstructured physician notes, and using it to effectively learn disease group characteristics for prediction. In this article, a deep neural network based model for predicting ICD9 disease groups from physician notes in the form of unstructured text is discussed.	./cache/cord-027303-20plzyqd.txt	./txt/cord-027303-20plzyqd.txt