Abstract: Electronic Health Records (EHRs) have seen a rapid increase in adoption during the last decade. The narrative prose contained in clinical notes is unstructured and unlocking its full potential has proved challenging. Many studies incorporating clinical notes have applied simple information extraction models to build representations that enhance a downstream clinical prediction task, such as mortality or readmission. Improved predictive performance suggests a "good" representation. However, these extrinsic evaluations are blind to most of the insight contained in the notes. In order to better understand the power of expressive clinical prose, we investigate both intrinsic and extrinsic methods for understanding several common note representations. To ensure replicability and to support the clinical modeling community, we run all experiments on publicly-available data and provide our code.
Learning Objective 1: After reading this paper, learners should be able to:
- describe several common clinical note representations, and
- understand that different representations capture different aspects of the underlying text.
Willie Boag (Presenter)
Dustin Doss, MIT
Tristan Naumann, MIT
Peter Szolovits, MIT