EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Identifying health conditions in older adults in textual health records using deep learning-based natural language processing

Jake Lin, Anna Kuukka, Tomi Korpi, Anna Tirkkonen, Antti Kariluoto, Juho Kaijansinkko, Maija Satamo, Hanna Pajulammi, Markus J Haapanen, Sergei Häyrynen, Eetu Pursiainen, Daniel Ciovica und 2 weitere

Computational and Structural Biotechnology Journal · 2025-01 · 28 · 341-347

DOI: 10.1016/j.csbj.2025.09.009 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Computational and Structural Biotechnology Journal, online erschienen am 12.09.2025. Verfasst von Jake Lin und 13 weiteren Autor:innen. Beteiligte Einrichtung: Faculty of Medicine and Health Technology, Tampere University, Tampere, Finland. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 25 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Sigrid Jusélius Foundation, Tampereen Yliopisto, Instrumentariumin Tiedesäätiö.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Many clinically significant health conditions in older adults are underreported or only recorded in unstructured health records. These records, however, contain valuable information for patient care and prognosis. This study utilized 10.6 million free-text entries from the electronic health records of 102,525 patients aged 50-80 across various care settings in Finland from 2010 to 2022. A deep learning-based natural language processing model was employed to perform named entity recognition (NER) to identify falls, incontinence, loneliness, and mobility limitations from the free-text entries. The performance of the NER models was evaluated by precision, recall and F1 scores. Diagnostic codes for incontinence and falls were collected for comparisons. Cox regression models were used to assess the predictive value of the identified conditions for all-cause mortality. The NER models demonstrated excellent performance with recall, precision and F1 scores greater than 0.80 across the health conditions. Compared to diagnostic codes, NER identified greater numbers of falls (31987 vs 4090) and incontinence (7059 vs 3873) onsets and yielded greater hazard ratios for all-cause mortality: 1.31 vs 1.04 for falls and 1.99 vs 0.65 for incontinence. Deep learning-based NER models present new opportunities to identify vulnerable patients in free text health records.

Schlagwörter

Stichwörter der Autor:innen

Electronic health recordsFallsIncontinenceLonelinessMobility limitationsNatural language processing

Autor:innen und Institutionen

  • Faculty of Medicine and Health Technology, Tampere University, Tampere, Finland.
  • Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Sweden.
  • Faculty of Sport and Health Sciences and Gerontology Research Center, University of Jyväskylä, Finland.
  • Faculty of Information Technology, University of Jyväskylä, Jyväskylä, Finland.
  • Department of Geriatric Medicine, Central Finland Hospital Nova, Wellbeing Services County of Central Finland, Finland.
  • Department of General Practice and Primary Health Care, University of Helsinki and Helsinki University Hospital, Helsinki, Finland.

Förderung

  • Sigrid Jusélius Foundation
  • Tampereen Yliopisto
  • Instrumentariumin Tiedesäätiö
  • Samfundet Folkhälsan
  • Research Council of Finland

Zitieren

APA

Lin, J., Kuukka, A., Korpi, T., Tirkkonen, A., Kariluoto, A., Kaijansinkko, J., Satamo, M., Pajulammi, H., Haapanen, M. J., Häyrynen, S., Pursiainen, E., Ciovica, D., von Bonsdorff, M. B., & Jylhävä, J. (2025). Identifying health conditions in older adults in textual health records using deep learning-based natural language processing. Computational and Structural Biotechnology Journal, 28, 341-347. https://doi.org/10.1016/j.csbj.2025.09.009

Vancouver

Lin J, Kuukka A, Korpi T, Tirkkonen A, Kariluoto A, Kaijansinkko J, et al. Identifying health conditions in older adults in textual health records using deep learning-based natural language processing. Computational and Structural Biotechnology Journal. 2025;28:341-347. doi:10.1016/j.csbj.2025.09.009

BibTeX
@article{Lin2025,
  author = {Lin, Jake and Kuukka, Anna and Korpi, Tomi and Tirkkonen, Anna and Kariluoto, Antti and Kaijansinkko, Juho and Satamo, Maija and Pajulammi, Hanna and Haapanen, Markus J and Häyrynen, Sergei and Pursiainen, Eetu and Ciovica, Daniel and von Bonsdorff, Mikaela B and Jylhävä, Juulia},
  title = {Identifying health conditions in older adults in textual health records using deep learning-based natural language processing},
  journal = {Computational and Structural Biotechnology Journal},
  year = {2025},
  volume = {28},
  pages = {341--347},
  doi = {10.1016/j.csbj.2025.09.009}
}
Kostenlos starten →