Learning to Ask Like a Physician

Existing question answering (QA) datasets derived from electronic health records (EHR) are artificially generated and consequently fail to capture realistic physician information needs. We present Discharge Summary Clinical Questions (DiSCQ), a newly curated question dataset composed of 2,000+ quest...

Full description

Saved in:

Bibliographic Details
Main Authors:	Lehman, Eric, Lialin, Vladislav, Legaspi, Katelyn Y, Sy, Anne Janelle R, Pile, Patricia Therese S, Alberto, Nicole Rose I, Ragasa, Richard Raymund R, Puyat, Corinna Victoria M, Alberto, Isabelle Rose I, Alfonso, Pia Gabrielle I, Taliño, Marianne, Moukheiber, Dana, Wallace, Byron C, Rumshisky, Anna, Liang, Jenifer J, Raghavan, Preethi, Celi, Leo Anthony, Szolovits, Peter
Format:	Journal Article
Language:	English
Published:	06-06-2022
Subjects:	Computer Science - Computation and Language
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	Existing question answering (QA) datasets derived from electronic health records (EHR) are artificially generated and consequently fail to capture realistic physician information needs. We present Discharge Summary Clinical Questions (DiSCQ), a newly curated question dataset composed of 2,000+ questions paired with the snippets of text (triggers) that prompted each question. The questions are generated by medical experts from 100+ MIMIC-III discharge summaries. We analyze this dataset to characterize the types of information sought by medical experts. We also train baseline models for trigger detection and question generation (QG), paired with unsupervised answer retrieval over EHRs. Our baseline model is able to generate high quality questions in over 62% of cases when prompted with human selected triggers. We release this dataset (and all code to reproduce baseline model results) to facilitate further research into realistic clinical QA and QG: https://github.com/elehman16/discq.
DOI:	10.48550/arxiv.2206.02696