Visual summarisation of text for surveillance and situational awareness in hospitals

Hanna Suominen, Leif Hanlen

Research output: A Conference proceeding or a Chapter in BookConference contribution

Abstract

Nosocomial infections (NIs, any infection that a patient contracts in a healthcare institution) cost 100; 000 lives and five billion dollars per year for 300 million Americans alone. Surveillance in hospitals holds the potential of reducing NI rates by more than thirty per cent but performing this task by hand is impossible at scale of every appointment, examination, intervention, and other event in healthcare. Narratives in patient records can indicate NIs and their automated processing could scale out surveillance. This paper describes a text summarisation system for NI surveillance and situational awareness in hospitals. The system is a cascaded sentence, report, and patient classifier. It generates three types of visual summaries for an input of patient narratives and ward maps: cross-sectional statuses at the same point of time, longitudinal trends in time, and highlighted text to see the textual evidence leading to a given status or trend. This gives evidence for and against a given NI in the levels of hospitals, wards, patients, reports, and sentences. The system has excellent recall and precision (e.g., 0.95 and 0.71 for reports) in summarisation for the subset of NIs from fungal species on 1; 880 authentic records of 527 patients from 3 hospitals. To demonstrate the system design, we have developed a mobile iPad compatible web-application and a simulation with eighteen patients on three medical wards in one hospital during one month with 61 records in total. The design is extendable to other summarisation tasks.

Original languageEnglish
Title of host publicationADCS 2013 - Proceedings of the 18th Australasian Document Computing Symposium
EditorsHanna Suominen, Leif Hanlen
Place of PublicationNew York
PublisherAssociation for Computing Machinery (ACM)
Pages89-96
Number of pages8
ISBN (Print)9781450325240
DOIs
Publication statusPublished - 5 Dec 2013
Event18th Australasian Document Computing Symposium, ADCS 2013 - Brisbane, Brisbane, Australia
Duration: 5 Dec 20136 Dec 2013

Conference

Conference18th Australasian Document Computing Symposium, ADCS 2013
Abbreviated titleADCS
CountryAustralia
CityBrisbane
Period5/12/136/12/13

Fingerprint

Classifiers
Systems analysis
Processing
Costs

Cite this

Suominen, H., & Hanlen, L. (2013). Visual summarisation of text for surveillance and situational awareness in hospitals. In H. Suominen, & L. Hanlen (Eds.), ADCS 2013 - Proceedings of the 18th Australasian Document Computing Symposium (pp. 89-96). New York: Association for Computing Machinery (ACM). https://doi.org/10.1145/2537734.2537739
Suominen, Hanna ; Hanlen, Leif. / Visual summarisation of text for surveillance and situational awareness in hospitals. ADCS 2013 - Proceedings of the 18th Australasian Document Computing Symposium. editor / Hanna Suominen ; Leif Hanlen. New York : Association for Computing Machinery (ACM), 2013. pp. 89-96
@inproceedings{cc12c408ea98413c9b07b0b9808bc88a,
title = "Visual summarisation of text for surveillance and situational awareness in hospitals",
abstract = "Nosocomial infections (NIs, any infection that a patient contracts in a healthcare institution) cost 100; 000 lives and five billion dollars per year for 300 million Americans alone. Surveillance in hospitals holds the potential of reducing NI rates by more than thirty per cent but performing this task by hand is impossible at scale of every appointment, examination, intervention, and other event in healthcare. Narratives in patient records can indicate NIs and their automated processing could scale out surveillance. This paper describes a text summarisation system for NI surveillance and situational awareness in hospitals. The system is a cascaded sentence, report, and patient classifier. It generates three types of visual summaries for an input of patient narratives and ward maps: cross-sectional statuses at the same point of time, longitudinal trends in time, and highlighted text to see the textual evidence leading to a given status or trend. This gives evidence for and against a given NI in the levels of hospitals, wards, patients, reports, and sentences. The system has excellent recall and precision (e.g., 0.95 and 0.71 for reports) in summarisation for the subset of NIs from fungal species on 1; 880 authentic records of 527 patients from 3 hospitals. To demonstrate the system design, we have developed a mobile iPad compatible web-application and a simulation with eighteen patients on three medical wards in one hospital during one month with 61 records in total. The design is extendable to other summarisation tasks.",
keywords = "Information storage and retrieval, Medical informatics, Medical records, Nosocomial infections",
author = "Hanna Suominen and Leif Hanlen",
year = "2013",
month = "12",
day = "5",
doi = "10.1145/2537734.2537739",
language = "English",
isbn = "9781450325240",
pages = "89--96",
editor = "Hanna Suominen and Leif Hanlen",
booktitle = "ADCS 2013 - Proceedings of the 18th Australasian Document Computing Symposium",
publisher = "Association for Computing Machinery (ACM)",
address = "United States",

}

Suominen, H & Hanlen, L 2013, Visual summarisation of text for surveillance and situational awareness in hospitals. in H Suominen & L Hanlen (eds), ADCS 2013 - Proceedings of the 18th Australasian Document Computing Symposium. Association for Computing Machinery (ACM), New York, pp. 89-96, 18th Australasian Document Computing Symposium, ADCS 2013, Brisbane, Australia, 5/12/13. https://doi.org/10.1145/2537734.2537739

Visual summarisation of text for surveillance and situational awareness in hospitals. / Suominen, Hanna; Hanlen, Leif.

ADCS 2013 - Proceedings of the 18th Australasian Document Computing Symposium. ed. / Hanna Suominen; Leif Hanlen. New York : Association for Computing Machinery (ACM), 2013. p. 89-96.

Research output: A Conference proceeding or a Chapter in BookConference contribution

TY - GEN

T1 - Visual summarisation of text for surveillance and situational awareness in hospitals

AU - Suominen, Hanna

AU - Hanlen, Leif

PY - 2013/12/5

Y1 - 2013/12/5

N2 - Nosocomial infections (NIs, any infection that a patient contracts in a healthcare institution) cost 100; 000 lives and five billion dollars per year for 300 million Americans alone. Surveillance in hospitals holds the potential of reducing NI rates by more than thirty per cent but performing this task by hand is impossible at scale of every appointment, examination, intervention, and other event in healthcare. Narratives in patient records can indicate NIs and their automated processing could scale out surveillance. This paper describes a text summarisation system for NI surveillance and situational awareness in hospitals. The system is a cascaded sentence, report, and patient classifier. It generates three types of visual summaries for an input of patient narratives and ward maps: cross-sectional statuses at the same point of time, longitudinal trends in time, and highlighted text to see the textual evidence leading to a given status or trend. This gives evidence for and against a given NI in the levels of hospitals, wards, patients, reports, and sentences. The system has excellent recall and precision (e.g., 0.95 and 0.71 for reports) in summarisation for the subset of NIs from fungal species on 1; 880 authentic records of 527 patients from 3 hospitals. To demonstrate the system design, we have developed a mobile iPad compatible web-application and a simulation with eighteen patients on three medical wards in one hospital during one month with 61 records in total. The design is extendable to other summarisation tasks.

AB - Nosocomial infections (NIs, any infection that a patient contracts in a healthcare institution) cost 100; 000 lives and five billion dollars per year for 300 million Americans alone. Surveillance in hospitals holds the potential of reducing NI rates by more than thirty per cent but performing this task by hand is impossible at scale of every appointment, examination, intervention, and other event in healthcare. Narratives in patient records can indicate NIs and their automated processing could scale out surveillance. This paper describes a text summarisation system for NI surveillance and situational awareness in hospitals. The system is a cascaded sentence, report, and patient classifier. It generates three types of visual summaries for an input of patient narratives and ward maps: cross-sectional statuses at the same point of time, longitudinal trends in time, and highlighted text to see the textual evidence leading to a given status or trend. This gives evidence for and against a given NI in the levels of hospitals, wards, patients, reports, and sentences. The system has excellent recall and precision (e.g., 0.95 and 0.71 for reports) in summarisation for the subset of NIs from fungal species on 1; 880 authentic records of 527 patients from 3 hospitals. To demonstrate the system design, we have developed a mobile iPad compatible web-application and a simulation with eighteen patients on three medical wards in one hospital during one month with 61 records in total. The design is extendable to other summarisation tasks.

KW - Information storage and retrieval

KW - Medical informatics

KW - Medical records

KW - Nosocomial infections

UR - http://www.scopus.com/inward/record.url?scp=84892686675&partnerID=8YFLogxK

U2 - 10.1145/2537734.2537739

DO - 10.1145/2537734.2537739

M3 - Conference contribution

SN - 9781450325240

SP - 89

EP - 96

BT - ADCS 2013 - Proceedings of the 18th Australasian Document Computing Symposium

A2 - Suominen, Hanna

A2 - Hanlen, Leif

PB - Association for Computing Machinery (ACM)

CY - New York

ER -

Suominen H, Hanlen L. Visual summarisation of text for surveillance and situational awareness in hospitals. In Suominen H, Hanlen L, editors, ADCS 2013 - Proceedings of the 18th Australasian Document Computing Symposium. New York: Association for Computing Machinery (ACM). 2013. p. 89-96 https://doi.org/10.1145/2537734.2537739