EUVIMEDEUVIMED entdecken
Open Access

Semantic Similarity Search Approach to Extract Exemplars of Stigmatizing and Positive Language in Obstetric Clinical Notes: Exploratory Study

Jihye Kim Scroggins, Ismael Ibrahim Hulchafo, Veronica Barcelona, Anahita Davoudi, Hans Moen, Sarah Harkins, Danielle Scharp, Kenrick Cato, Michele Tadiello, Maxim Topaz

JMIR Medical Informatics · 22.09.2026 · 14 · e85088-e85088

DOI: 10.2196/85088 ↗

Kurz gefasst

Fachartikel in der Zeitschrift JMIR Medical Informatics, online erschienen am 22.09.2026. Verfasst von Jihye Kim Scroggins und 9 weiteren Autor:innen. Beteiligte Einrichtung: School of Nursing, University of North Carolina at Chapel Hill, 120 Medical Drive, Chapel Hill, NC, 27514, United States, 1 (919) 966-4260. Hauptthemen laut MEDLINE: Electronic Health Records, Natural Language Processing, Semantics, Obstetrics. Die Arbeit nennt 48 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Natural language processing can extract meaningful information from clinical notes. However, human annotation is time-consuming and costly, and scarce data poses a challenge.

Objective

This study aimed to explore a semantic similarity search approach to extract exemplars of stigmatizing and positive language in obstetric clinical notes.

Methods

We used electronic health record data from labor and birth admissions at 2 hospitals in the United States from 2017 to 2019. We used a semantic similarity search approach, which used 200 randomly selected true exemplars, stratified by language categories, as queries to search for similar exemplar candidates. We extracted the top 5 candidates with the highest cosine similarities, which were assessed for accuracy.

Results

We retrieved 1000 candidates. An average precision of 0.69 was achieved when candidates with cosine similarity thresholds of 0.75 or higher were included, at which point 68.8% (64/93) of exemplar candidates accurately represented true cases. At the 0.75 threshold, the proportion of true cases was higher for preferred language (41/56, 73.2%) and unilateral/authoritarian decisions (5/7, 71.4%). The proportion of true cases was lower for difficult patients (2/7, 28.6%) and marginalized identities (3/9, 33.3%).

Conclusions

The semantic similarity search approach shows promise in efficiently extracting exemplars while reducing the annotation burden, laying the groundwork for future applications in other domains.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Electronic Health RecordsNatural Language ProcessingSemanticsObstetricsStereotypingHumansFemalePregnancyUnited StatesLanguage

Stichwörter der Autor:innen

biaselectronic health recordshealth communicationnatural language processingstigmatizing language

Autor:innen und Institutionen

  • School of Nursing, University of North Carolina at Chapel Hill, 120 Medical Drive, Chapel Hill, NC, 27514, United States, 1 (919) 966-4260.
  • School of Nursing, Columbia University, New York, NY, United States.
  • VNS Health, New York, NY, United States.
  • Department of Computer Science, Aalto University, Espoo, Finland.
  • Icahn School of Medicine at Mount Sinai, New York, NY, United States.
  • School of Nursing, University of Pennsylvania, Philadelphia, PA, United States.

Zitieren

APA

Scroggins, J. K., Hulchafo, I. I., Barcelona, V., Davoudi, A., Moen, H., Harkins, S., Scharp, D., Cato, K., Tadiello, M., & Topaz, M. (2026). Semantic Similarity Search Approach to Extract Exemplars of Stigmatizing and Positive Language in Obstetric Clinical Notes: Exploratory Study. JMIR Medical Informatics, 14, e85088-e85088. https://doi.org/10.2196/85088

Vancouver

Scroggins JK, Hulchafo II, Barcelona V, Davoudi A, Moen H, Harkins S, et al. Semantic Similarity Search Approach to Extract Exemplars of Stigmatizing and Positive Language in Obstetric Clinical Notes: Exploratory Study. JMIR Medical Informatics. 2026;14:e85088-e85088. doi:10.2196/85088

BibTeX
@article{Scroggins2026,
  author = {Scroggins, Jihye Kim and Hulchafo, Ismael Ibrahim and Barcelona, Veronica and Davoudi, Anahita and Moen, Hans and Harkins, Sarah and Scharp, Danielle and Cato, Kenrick and Tadiello, Michele and Topaz, Maxim},
  title = {Semantic Similarity Search Approach to Extract Exemplars of Stigmatizing and Positive Language in Obstetric Clinical Notes: Exploratory Study},
  journal = {JMIR Medical Informatics},
  year = {2026},
  volume = {14},
  pages = {e85088--e85088},
  doi = {10.2196/85088}
}
Kostenlos starten →