Skip to content

Latest commit

 

History

History
9 lines (5 loc) · 696 Bytes

File metadata and controls

9 lines (5 loc) · 696 Bytes

enriched_LSTMs

Getting more out of LSTMs for classifying multimodal health data

Background

Our data

We used emergency department (ED) visit record data to develop our models. The records had one free-text field, chief complaint, along with a number of other discrete variables, like age group, sex, mode of arrival, and hospital code. This is the third project we've done with the data, so if you're interested in learning more about them, check out our papers about using them for classifying chief complaints and for generating synthetic chief complaints.