Extraction and Analysis of Clinically Important Follow-up Recommendations in a Large Radiology Dataset
©2020 AMIA - All rights reserved..
Communication of follow-up recommendations when abnormalities are identified on imaging studies is prone to error. In this paper, we present a natural language processing approach based on deep learning to automatically identify clinically important recommendations in radiology reports. Our approach first identifies the recommendation sentences and then extracts reason, test, and time frame of the identified recommendations. To train our extraction models, we created a corpus of 1367 radiology reports annotated for recommendation information. Our extraction models achieved 0.93 f-score for recommendation sentence, 0.65 f-score for reason, 0.73 f-score for test, and 0.84 f-score for time frame. We applied the extraction models to a set of over 3.3 million radiology reports and analyzed the adherence of follow-up recommendations.
Medienart: |
Artikel |
---|
Erscheinungsjahr: |
2020 |
---|---|
Erschienen: |
2020 |
Enthalten in: |
Zur Gesamtaufnahme - volume:2020 |
---|---|
Enthalten in: |
AMIA Joint Summits on Translational Science proceedings. AMIA Joint Summits on Translational Science - 2020(2020) vom: 22., Seite 335-344 |
Sprache: |
Englisch |
---|
Beteiligte Personen: |
Lau, Wilson [VerfasserIn] |
---|
Themen: |
---|
Anmerkungen: |
Date Revised 28.09.2020 published: Electronic-eCollection Citation Status PubMed-not-MEDLINE |
---|
Förderinstitution / Projekttitel: |
|
---|
PPN (Katalog-ID): |
NLM310580846 |
---|
LEADER | 01000caa a22002652 4500 | ||
---|---|---|---|
001 | NLM310580846 | ||
003 | DE-627 | ||
005 | 20231226201507.0 | ||
007 | tu | ||
008 | 231225s2020 xx ||||| 00| ||eng c | ||
028 | 5 | 2 | |a pubmed24n1035.xml |
035 | |a (DE-627)NLM310580846 | ||
035 | |a (NLM)32477653 | ||
040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
041 | |a eng | ||
100 | 1 | |a Lau, Wilson |e verfasserin |4 aut | |
245 | 1 | 0 | |a Extraction and Analysis of Clinically Important Follow-up Recommendations in a Large Radiology Dataset |
264 | 1 | |c 2020 | |
336 | |a Text |b txt |2 rdacontent | ||
337 | |a ohne Hilfsmittel zu benutzen |b n |2 rdamedia | ||
338 | |a Band |b nc |2 rdacarrier | ||
500 | |a Date Revised 28.09.2020 | ||
500 | |a published: Electronic-eCollection | ||
500 | |a Citation Status PubMed-not-MEDLINE | ||
520 | |a ©2020 AMIA - All rights reserved. | ||
520 | |a Communication of follow-up recommendations when abnormalities are identified on imaging studies is prone to error. In this paper, we present a natural language processing approach based on deep learning to automatically identify clinically important recommendations in radiology reports. Our approach first identifies the recommendation sentences and then extracts reason, test, and time frame of the identified recommendations. To train our extraction models, we created a corpus of 1367 radiology reports annotated for recommendation information. Our extraction models achieved 0.93 f-score for recommendation sentence, 0.65 f-score for reason, 0.73 f-score for test, and 0.84 f-score for time frame. We applied the extraction models to a set of over 3.3 million radiology reports and analyzed the adherence of follow-up recommendations | ||
650 | 4 | |a Journal Article | |
700 | 1 | |a Payne, Thomas H |e verfasserin |4 aut | |
700 | 1 | |a Uzuner, Ozlem |e verfasserin |4 aut | |
700 | 1 | |a Yetisgen, Meliha |e verfasserin |4 aut | |
773 | 0 | 8 | |i Enthalten in |t AMIA Joint Summits on Translational Science proceedings. AMIA Joint Summits on Translational Science |d 2012 |g 2020(2020) vom: 22., Seite 335-344 |w (DE-627)NLM219295107 |x 2153-4063 |7 nnns |
773 | 1 | 8 | |g volume:2020 |g year:2020 |g day:22 |g pages:335-344 |
912 | |a GBV_USEFLAG_A | ||
912 | |a GBV_NLM | ||
951 | |a AR | ||
952 | |d 2020 |j 2020 |b 22 |h 335-344 |