Towards automated classification of intensive care nursing narratives

Marketta Hiissa, Tapio Pahikkala, Hanna Suominen, Tuija Lehtikunnas, Barbro Back, Helena Karsten, Sanna Salanterä, Tapio I. Salakoski

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

Background: Nursing narratives are an important part of patient documentation, but the possibilities to utilize them in the direct care process are limited due to the lack of proper tools. One solution to facilitate the utilization of narrative data could be to classify them according to their content. Objectives: Our objective is to address two issues related to designing an automated classifier: domain experts' agreement on the content of classes Breathing, Blood Circulation and Pain, as well as the ability of a machine-learning-based classifier to learn the classification patterns of the nurses. Methods: The data we used were a set of Finnish intensive care nursing narratives, and we used the regularized least-squares (RLS) algorithm for the automatic classification. The agreement of the nurses was assessed by using Cohen's κ, and the performance of the algorithm was measured using area under ROC curve (AUC). Results: On average, the values of κ were around 0.8. The agreement was highest in the class Blood Circulation, and lowest in the class Breathing. The RLS algorithm was able to learn the classification patterns of the three nurses on an acceptable level; the values of AUC were generally around 0.85. Conclusions: Our results indicate that the free text in nursing documentation can be automatically classified and this can offer a way to develop electronic patient records.

Original languageEnglish
Pages (from-to)S362-S368
Number of pages7
JournalInternational Journal of Medical Informatics
Volume76
Issue numberSUPPL. 3
DOIs
Publication statusPublished - Dec 2007
Externally publishedYes

Fingerprint

Critical Care Nursing
Blood Circulation
Nurses
Least-Squares Analysis
ROC Curve
Documentation
Area Under Curve
Respiration
Nursing
Aptitude
Pain

Cite this

Hiissa, M., Pahikkala, T., Suominen, H., Lehtikunnas, T., Back, B., Karsten, H., ... Salakoski, T. I. (2007). Towards automated classification of intensive care nursing narratives. International Journal of Medical Informatics, 76(SUPPL. 3), S362-S368. https://doi.org/10.1016/j.ijmedinf.2007.03.003
Hiissa, Marketta ; Pahikkala, Tapio ; Suominen, Hanna ; Lehtikunnas, Tuija ; Back, Barbro ; Karsten, Helena ; Salanterä, Sanna ; Salakoski, Tapio I. / Towards automated classification of intensive care nursing narratives. In: International Journal of Medical Informatics. 2007 ; Vol. 76, No. SUPPL. 3. pp. S362-S368.
@article{49ee8bbae7674f63a07186be19e20626,
title = "Towards automated classification of intensive care nursing narratives",
abstract = "Background: Nursing narratives are an important part of patient documentation, but the possibilities to utilize them in the direct care process are limited due to the lack of proper tools. One solution to facilitate the utilization of narrative data could be to classify them according to their content. Objectives: Our objective is to address two issues related to designing an automated classifier: domain experts' agreement on the content of classes Breathing, Blood Circulation and Pain, as well as the ability of a machine-learning-based classifier to learn the classification patterns of the nurses. Methods: The data we used were a set of Finnish intensive care nursing narratives, and we used the regularized least-squares (RLS) algorithm for the automatic classification. The agreement of the nurses was assessed by using Cohen's κ, and the performance of the algorithm was measured using area under ROC curve (AUC). Results: On average, the values of κ were around 0.8. The agreement was highest in the class Blood Circulation, and lowest in the class Breathing. The RLS algorithm was able to learn the classification patterns of the three nurses on an acceptable level; the values of AUC were generally around 0.85. Conclusions: Our results indicate that the free text in nursing documentation can be automatically classified and this can offer a way to develop electronic patient records.",
keywords = "Computerized patient records, Intensive care, Natural language processing, Nursing, Nursing records",
author = "Marketta Hiissa and Tapio Pahikkala and Hanna Suominen and Tuija Lehtikunnas and Barbro Back and Helena Karsten and Sanna Salanter{\"a} and Salakoski, {Tapio I.}",
year = "2007",
month = "12",
doi = "10.1016/j.ijmedinf.2007.03.003",
language = "English",
volume = "76",
pages = "S362--S368",
journal = "International Journal of Bio-Medical Computing",
issn = "1386-5056",
publisher = "Elsevier Ireland Ltd",
number = "SUPPL. 3",

}

Hiissa, M, Pahikkala, T, Suominen, H, Lehtikunnas, T, Back, B, Karsten, H, Salanterä, S & Salakoski, TI 2007, 'Towards automated classification of intensive care nursing narratives', International Journal of Medical Informatics, vol. 76, no. SUPPL. 3, pp. S362-S368. https://doi.org/10.1016/j.ijmedinf.2007.03.003

Towards automated classification of intensive care nursing narratives. / Hiissa, Marketta; Pahikkala, Tapio; Suominen, Hanna; Lehtikunnas, Tuija; Back, Barbro; Karsten, Helena; Salanterä, Sanna; Salakoski, Tapio I.

In: International Journal of Medical Informatics, Vol. 76, No. SUPPL. 3, 12.2007, p. S362-S368.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Towards automated classification of intensive care nursing narratives

AU - Hiissa, Marketta

AU - Pahikkala, Tapio

AU - Suominen, Hanna

AU - Lehtikunnas, Tuija

AU - Back, Barbro

AU - Karsten, Helena

AU - Salanterä, Sanna

AU - Salakoski, Tapio I.

PY - 2007/12

Y1 - 2007/12

N2 - Background: Nursing narratives are an important part of patient documentation, but the possibilities to utilize them in the direct care process are limited due to the lack of proper tools. One solution to facilitate the utilization of narrative data could be to classify them according to their content. Objectives: Our objective is to address two issues related to designing an automated classifier: domain experts' agreement on the content of classes Breathing, Blood Circulation and Pain, as well as the ability of a machine-learning-based classifier to learn the classification patterns of the nurses. Methods: The data we used were a set of Finnish intensive care nursing narratives, and we used the regularized least-squares (RLS) algorithm for the automatic classification. The agreement of the nurses was assessed by using Cohen's κ, and the performance of the algorithm was measured using area under ROC curve (AUC). Results: On average, the values of κ were around 0.8. The agreement was highest in the class Blood Circulation, and lowest in the class Breathing. The RLS algorithm was able to learn the classification patterns of the three nurses on an acceptable level; the values of AUC were generally around 0.85. Conclusions: Our results indicate that the free text in nursing documentation can be automatically classified and this can offer a way to develop electronic patient records.

AB - Background: Nursing narratives are an important part of patient documentation, but the possibilities to utilize them in the direct care process are limited due to the lack of proper tools. One solution to facilitate the utilization of narrative data could be to classify them according to their content. Objectives: Our objective is to address two issues related to designing an automated classifier: domain experts' agreement on the content of classes Breathing, Blood Circulation and Pain, as well as the ability of a machine-learning-based classifier to learn the classification patterns of the nurses. Methods: The data we used were a set of Finnish intensive care nursing narratives, and we used the regularized least-squares (RLS) algorithm for the automatic classification. The agreement of the nurses was assessed by using Cohen's κ, and the performance of the algorithm was measured using area under ROC curve (AUC). Results: On average, the values of κ were around 0.8. The agreement was highest in the class Blood Circulation, and lowest in the class Breathing. The RLS algorithm was able to learn the classification patterns of the three nurses on an acceptable level; the values of AUC were generally around 0.85. Conclusions: Our results indicate that the free text in nursing documentation can be automatically classified and this can offer a way to develop electronic patient records.

KW - Computerized patient records

KW - Intensive care

KW - Natural language processing

KW - Nursing

KW - Nursing records

UR - http://www.scopus.com/inward/record.url?scp=34548686982&partnerID=8YFLogxK

U2 - 10.1016/j.ijmedinf.2007.03.003

DO - 10.1016/j.ijmedinf.2007.03.003

M3 - Article

VL - 76

SP - S362-S368

JO - International Journal of Bio-Medical Computing

JF - International Journal of Bio-Medical Computing

SN - 1386-5056

IS - SUPPL. 3

ER -

Hiissa M, Pahikkala T, Suominen H, Lehtikunnas T, Back B, Karsten H et al. Towards automated classification of intensive care nursing narratives. International Journal of Medical Informatics. 2007 Dec;76(SUPPL. 3):S362-S368. https://doi.org/10.1016/j.ijmedinf.2007.03.003