Learning to Ask Like a Physician
Name
lehman-lehmer16-SM-EECS-2022-thesis.pdf
Description
Thesis PDF
Size
717.56 KB
Format
Adobe PDF
Checksum (MD5)
47ffcfcae5e9ef64fbc4110e7687eb06
Author(s)
Lehman, Eric(Computer scientist)
Advisor(s)
Szolovits, Peter
Date Issued
May 2022
Publisher
Massachusetts Institute of Technology
Abstract
Existing question answering (QA) datasets derived from electronic health records (EHR) are artificially generated and, as a result, 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 will release this dataset (and all code to reproduce baseline model results) to facilitate further research into realistic clinical QA and QG.
MIT Department
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
Terms of Use
In Copyright - Educational Use Permitted
Copyright MIT
Persistent DSpace Link
