EUVIMEDObjevte EUVIMED
BeobachtungsstudieOpen access

Automated Identification of Surgical Site Infections From Electronic Medical Records: Retrospective Observational Predictive Modeling Study

Arjun Chakraborty iD, Peter Tarczy-Hornoch iD, Dustin Long iD, Meliha Yetisgen iD

JMIR Perioperative Medicine · 2026-06-26 · 9 · e87896-e87896

DOI: 10.2196/87896 ↗

At a glance

Beobachtungsstudie in JMIR Perioperative Medicine, published online on 2026-06-26. Written by Arjun Chakraborty and 3 co-authors. Institution: Department of Biomedical Informatics and Medical Education, School of Medicine, University of Washington, 222 15th Street SW, Rochester, MN, 55902, U…. Major MEDLINE topics: Surgical Wound Infection, Electronic Health Records. 27 references. The full text is freely available in PubMed Central. Funded by: NIAMS NIH HHS.

AbstraktSubjectsBibliografické údajeCiteRelated studies

Abstract (original language)

Background

Surgical site infections (SSIs) affect 160,000 to 300,000 patients annually, increasing postoperative mortality, causing significant complications, and incurring US $3.5 to US $10 billion in excess costs each year. Effective SSI surveillance can inform strategies to mitigate these outcomes. Traditional SSI surveillance methods, primarily manual chart reviews, are costly and labor-intensive.

Objective

This study aimed to evaluate whether an automated SSI surveillance system built using newer natural language processing methods and deep learning could outperform previous approaches and whether such an approach could enable more efficient infection surveillance.

Methods

Our dataset comprised approximately 30,000 surgical cases from the University of Washington Medical Center (UWMC) and Harborview Medical Center (HMC). Data from UWMC were captured for the National Surgical Quality Improvement Program, and data from HMC were captured for the National Healthcare Safety Network. Electronic health record (EHR) data for each surgical case included structured data pertaining to surgical procedure characteristics, laboratory values, and antibiotic administration, as well as clinical text notes for a surgical case from 7 days before to 90 days after surgery. Using this data, we applied a myriad of machine learning approaches to the task of SSI prediction. We reported the following performance metrics: F 1 -score, precision (positive predictive value), recall (sensitivity), area under the precision-recall curve, and precision at 0.9 recall for each machine learning approach.

Results

In a cohort of 5996 surgical cases, incorporating multimodal EHR information—including contextual information from clinical text and temporal information from laboratory values—improved SSI prediction performance. Models using structured data and clinical text outperformed structured data alone ( F 1 =0.68, 95% CI 0.68‐0.69 vs F 1 =0.55, 95% CI 0.54‐0.56; P

Subjects

MeSH terms (MEDLINE)

Surgical Wound InfectionElectronic Health RecordsHumansRetrospective StudiesNatural Language ProcessingPredictive Learning Models

Author keywords

deep learningdiagnosiselectronic health recordepidemiologylarge language modelsmachine learningnatural language processingneural networks computersurgical wound infection

Authors and institutions

  • Department of Biomedical Informatics and Medical Education, School of Medicine, University of Washington, 222 15th Street SW, Rochester, MN, 55902, United States, 1 5107095904.
  • Department of Pediatrics, School of Medicine, University of Washington, Seattle, WA, United States.
  • Department of Computer Science and Engineering, Paul G. Allen School of Computer Science and Engineering, College of Engineering, University of Washington, Seattle, WA, United States.
  • Department of Anesthesiology and Pain Medicine, School of Medicine, University of Washington, Seattle, WA, United States.

ORCID profiles: Arjun Chakraborty, Peter Tarczy-Hornoch, Dustin Long, Meliha Yetisgen

Funding

  • NIAMS NIH HHS

Cite

APA

Chakraborty, A., Tarczy-Hornoch, P., Long, D., & Yetisgen, M. (2026). Automated Identification of Surgical Site Infections From Electronic Medical Records: Retrospective Observational Predictive Modeling Study. JMIR Perioperative Medicine, 9, e87896-e87896. https://doi.org/10.2196/87896

Vancouver

Chakraborty A, Tarczy-Hornoch P, Long D, Yetisgen M. Automated Identification of Surgical Site Infections From Electronic Medical Records: Retrospective Observational Predictive Modeling Study. JMIR Perioperative Medicine. 2026;9:e87896-e87896. doi:10.2196/87896

BibTeX
@article{Chakraborty2026,
  author = {Chakraborty, Arjun and Tarczy-Hornoch, Peter and Long, Dustin and Yetisgen, Meliha},
  title = {Automated Identification of Surgical Site Infections From Electronic Medical Records: Retrospective Observational Predictive Modeling Study},
  journal = {JMIR Perioperative Medicine},
  year = {2026},
  volume = {9},
  pages = {e87896--e87896},
  doi = {10.2196/87896}
}
Start for free →