Discharge summary hospital course summarisation of in patient Electronic Health Record text with clinical concept guided deep pre-trained Transformer models

Thomas Searle; Zina Ibrahim; James Teo; Richard J.B. Dobson

doi:10.1016/j.jbi.2023.104358

Discharge summary hospital course summarisation of in patient Electronic Health Record text with clinical concept guided deep pre-trained Transformer models

Thomas Searle^*, Zina Ibrahim, James Teo, Richard J.B. Dobson

^*Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

2 Citations (Scopus)

Abstract

Brief Hospital Course (BHC) summaries are succinct summaries of an entire hospital encounter, embedded within discharge summaries, written by senior clinicians responsible for the overall care of a patient. Methods to automatically produce summaries from inpatient documentation would be invaluable in reducing clinician manual burden of summarising documents under high time-pressure to admit and discharge patients. Automatically producing these summaries from the inpatient course, is a complex, multi-document summarisation task, as source notes are written from various perspectives (e.g. nursing, doctor, radiology), during the course of the hospitalisation. We demonstrate a range of methods for BHC summarisation demonstrating the performance of deep learning summarisation models across extractive and abstractive summarisation scenarios. We also test a novel ensemble extractive and abstractive summarisation model that incorporates a medical concept ontology (SNOMED) as a clinical guidance signal and shows superior performance in 2 real-world clinical data sets.

Original language	English
Article number	104358
Journal	JOURNAL OF BIOMEDICAL INFORMATICS
Volume	141
DOIs	https://doi.org/10.1016/j.jbi.2023.104358
Publication status	Published - May 2023

Keywords

Clinical natural language processing
Clinical text summarisation
Pre-trained, deep learning, fine-tuned models for clinical summarisation

Access to Document

10.1016/j.jbi.2023.104358

Cite this

@article{3134b2c7a43d4e889ef46fe32db604a9,

title = "Discharge summary hospital course summarisation of in patient Electronic Health Record text with clinical concept guided deep pre-trained Transformer models",

abstract = "Brief Hospital Course (BHC) summaries are succinct summaries of an entire hospital encounter, embedded within discharge summaries, written by senior clinicians responsible for the overall care of a patient. Methods to automatically produce summaries from inpatient documentation would be invaluable in reducing clinician manual burden of summarising documents under high time-pressure to admit and discharge patients. Automatically producing these summaries from the inpatient course, is a complex, multi-document summarisation task, as source notes are written from various perspectives (e.g. nursing, doctor, radiology), during the course of the hospitalisation. We demonstrate a range of methods for BHC summarisation demonstrating the performance of deep learning summarisation models across extractive and abstractive summarisation scenarios. We also test a novel ensemble extractive and abstractive summarisation model that incorporates a medical concept ontology (SNOMED) as a clinical guidance signal and shows superior performance in 2 real-world clinical data sets.",

keywords = "Clinical natural language processing, Clinical text summarisation, Pre-trained, deep learning, fine-tuned models for clinical summarisation",

author = "Thomas Searle and Zina Ibrahim and James Teo and Dobson, {Richard J.B.}",

note = "Funding Information: RD{\textquoteright}s work is supported by 1. National Institute for Health Research (NIHR) Biomedical Research Centre at South London and Maudsley NHS Foundation Trust and King{\textquoteright}s College London . 2. Health Data Research UK, which is funded by the UK Medical Research Council , Engineering and Physical Sciences Research Council, United Kingdom , Economic and Social Research Council, United Kingdom , Department of Health and Social Care (England) , Chief Scientist Office of the Scottish Government Health and Social Care Directorates , Health and Social Care Research and Development Division (Welsh Government) , Public Health Agency (Northern Ireland) , British Heart Foundation and Wellcome Trust . 3. The National Institute for Health Research University College London Hospitals Biomedical Research Centre . This paper represents independent research part funded by the National Institute for Health Research (NIHR) Biomedical Research Centre at South London and Maudsley NHS Foundation Trust and King{\textquoteright}s College London . The views expressed are those of the author(s) and not necessarily those of the NHS, MRC, NIHR or the Department of Health and Social Care. Funding Information: RD's work is supported by 1. National Institute for Health Research (NIHR) Biomedical Research Centre at South London and Maudsley NHS Foundation Trust and King's College London. 2. Health Data Research UK, which is funded by the UK Medical Research Council, Engineering and Physical Sciences Research Council, United Kingdom, Economic and Social Research Council, United Kingdom, Department of Health and Social Care (England), Chief Scientist Office of the Scottish Government Health and Social Care Directorates, Health and Social Care Research and Development Division (Welsh Government), Public Health Agency (Northern Ireland), British Heart Foundation and Wellcome Trust. 3. The National Institute for Health Research University College London Hospitals Biomedical Research Centre. This paper represents independent research part funded by the National Institute for Health Research (NIHR) Biomedical Research Centre at South London and Maudsley NHS Foundation Trust and King's College London. The views expressed are those of the author(s) and not necessarily those of the NHS, MRC, NIHR or the Department of Health and Social Care. Publisher Copyright: {\textcopyright} 2023",

year = "2023",

month = may,

doi = "10.1016/j.jbi.2023.104358",

language = "English",

volume = "141",

journal = "JOURNAL OF BIOMEDICAL INFORMATICS",

issn = "1532-0464",

publisher = "ACADEMIC PRESS INC",

}

TY - JOUR

T1 - Discharge summary hospital course summarisation of in patient Electronic Health Record text with clinical concept guided deep pre-trained Transformer models

AU - Searle, Thomas

AU - Ibrahim, Zina

AU - Teo, James

AU - Dobson, Richard J.B.

N1 - Funding Information: RD’s work is supported by 1. National Institute for Health Research (NIHR) Biomedical Research Centre at South London and Maudsley NHS Foundation Trust and King’s College London . 2. Health Data Research UK, which is funded by the UK Medical Research Council , Engineering and Physical Sciences Research Council, United Kingdom , Economic and Social Research Council, United Kingdom , Department of Health and Social Care (England) , Chief Scientist Office of the Scottish Government Health and Social Care Directorates , Health and Social Care Research and Development Division (Welsh Government) , Public Health Agency (Northern Ireland) , British Heart Foundation and Wellcome Trust . 3. The National Institute for Health Research University College London Hospitals Biomedical Research Centre . This paper represents independent research part funded by the National Institute for Health Research (NIHR) Biomedical Research Centre at South London and Maudsley NHS Foundation Trust and King’s College London . The views expressed are those of the author(s) and not necessarily those of the NHS, MRC, NIHR or the Department of Health and Social Care. Funding Information: RD's work is supported by 1. National Institute for Health Research (NIHR) Biomedical Research Centre at South London and Maudsley NHS Foundation Trust and King's College London. 2. Health Data Research UK, which is funded by the UK Medical Research Council, Engineering and Physical Sciences Research Council, United Kingdom, Economic and Social Research Council, United Kingdom, Department of Health and Social Care (England), Chief Scientist Office of the Scottish Government Health and Social Care Directorates, Health and Social Care Research and Development Division (Welsh Government), Public Health Agency (Northern Ireland), British Heart Foundation and Wellcome Trust. 3. The National Institute for Health Research University College London Hospitals Biomedical Research Centre. This paper represents independent research part funded by the National Institute for Health Research (NIHR) Biomedical Research Centre at South London and Maudsley NHS Foundation Trust and King's College London. The views expressed are those of the author(s) and not necessarily those of the NHS, MRC, NIHR or the Department of Health and Social Care. Publisher Copyright: © 2023

PY - 2023/5

Y1 - 2023/5

N2 - Brief Hospital Course (BHC) summaries are succinct summaries of an entire hospital encounter, embedded within discharge summaries, written by senior clinicians responsible for the overall care of a patient. Methods to automatically produce summaries from inpatient documentation would be invaluable in reducing clinician manual burden of summarising documents under high time-pressure to admit and discharge patients. Automatically producing these summaries from the inpatient course, is a complex, multi-document summarisation task, as source notes are written from various perspectives (e.g. nursing, doctor, radiology), during the course of the hospitalisation. We demonstrate a range of methods for BHC summarisation demonstrating the performance of deep learning summarisation models across extractive and abstractive summarisation scenarios. We also test a novel ensemble extractive and abstractive summarisation model that incorporates a medical concept ontology (SNOMED) as a clinical guidance signal and shows superior performance in 2 real-world clinical data sets.

AB - Brief Hospital Course (BHC) summaries are succinct summaries of an entire hospital encounter, embedded within discharge summaries, written by senior clinicians responsible for the overall care of a patient. Methods to automatically produce summaries from inpatient documentation would be invaluable in reducing clinician manual burden of summarising documents under high time-pressure to admit and discharge patients. Automatically producing these summaries from the inpatient course, is a complex, multi-document summarisation task, as source notes are written from various perspectives (e.g. nursing, doctor, radiology), during the course of the hospitalisation. We demonstrate a range of methods for BHC summarisation demonstrating the performance of deep learning summarisation models across extractive and abstractive summarisation scenarios. We also test a novel ensemble extractive and abstractive summarisation model that incorporates a medical concept ontology (SNOMED) as a clinical guidance signal and shows superior performance in 2 real-world clinical data sets.

KW - Clinical natural language processing

KW - Clinical text summarisation

KW - Pre-trained, deep learning, fine-tuned models for clinical summarisation

UR - http://www.scopus.com/inward/record.url?scp=85152096026&partnerID=8YFLogxK

U2 - 10.1016/j.jbi.2023.104358

DO - 10.1016/j.jbi.2023.104358

M3 - Article

C2 - 37023846

AN - SCOPUS:85152096026

SN - 1532-0464

VL - 141

JO - JOURNAL OF BIOMEDICAL INFORMATICS

JF - JOURNAL OF BIOMEDICAL INFORMATICS

M1 - 104358

ER -

Discharge summary hospital course summarisation of in patient Electronic Health Record text with clinical concept guided deep pre-trained Transformer models

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this