Use of natural language processing to identify patients with inflammatory breast cancer across a health-care system
JNCI Cancer Spectrum · 2025
DOI: 10.1093/jncics/pkaf058 ↗Kurz gefasst
Fachartikel in der Zeitschrift JNCI Cancer Spectrum, 2025. Verfasst von Ramez Kouzy und 5 weiteren Autor:innen. Beteiligte Einrichtung: Division of Radiation Oncology, Department of Radiation Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX, United States. Hauptthemen laut MEDLINE: Natural Language Processing, Electronic Health Records, Inflammatory Breast Neoplasms, Early Detection of Cancer. Die Arbeit nennt 6 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: The University of Texas.
Abstract (Originalsprache)
Early identification and referral of inflammatory breast cancer remains challenging within large health-care systems, limiting access to specialized care. We developed and evaluated an artificial intelligence–driven platform integrating natural language processing (NLP) with electronic health records to systematically identify potential inflammatory breast cancer patients across 5 campuses. Our platform analyzed 8 623 494 clinical notes, implementing a sequential review process: NLP screening followed by human validation and multidisciplinary confirmation. Initial NLP screening achieved 55.4% positive predictive value, improving to 78.4% with human-in-the-loop review. Notably, among 255 confirmed patients with inflammatory breast cancer, our system demonstrated 92.2% sensitivity, identifying 57 patients (22.4%) that traditional surveillance methods missed. Documentation patterns influenced system performance, with combined inflammatory breast cancer and T4d staging mentions showing the highest predictive value (98.2%). This proof-of-concept study demonstrates that lightweight NLP systems with targeted human review can identify rare cancer cases that may otherwise remain siloed within complex health-care networks, ultimately improving access to specialized care resources.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Division of Radiation Oncology, Department of Radiation Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX, United States.
- Department of Breast Medical Oncology, Division of Cancer Medicine, The University of Texas MD Anderson Cancer Center, Houston, TX, United States.
- Morgan Welch Inflammatory Breast Cancer Research Program and Clinic, The University of Texas MD Anderson Cancer Center, Houston, TX, United States.
- Department of Breast Imaging, Division of Diagnostic Imaging, The University of Texas MD Anderson Cancer Center, Houston, TX, United States.
- Department of Breast Radiation Oncology, Division of Radiation Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX, United States.
ORCID-Profile: Ramez Kouzy, Megumi Kai, Huong T Le-Petross, Sadia Saleem, Wendy A Woodward
Förderung
- The University of Texas
Zitieren
APA
Kouzy, R., Kai, M., Le-Petross, H. T., Saleem, S., MDACC Inflammatory Breast Cancer Team, & Woodward, W. A. (2025). Use of natural language processing to identify patients with inflammatory breast cancer across a health-care system. JNCI Cancer Spectrum. https://doi.org/10.1093/jncics/pkaf058
Vancouver
Kouzy R, Kai M, Le-Petross HT, Saleem S, MDACC Inflammatory Breast Cancer Team, Woodward WA. Use of natural language processing to identify patients with inflammatory breast cancer across a health-care system. JNCI Cancer Spectrum. 2025. doi:10.1093/jncics/pkaf058
BibTeX
@article{Kouzy2025,
author = {Kouzy, Ramez and Kai, Megumi and Le-Petross, Huong T and Saleem, Sadia and MDACC Inflammatory Breast Cancer Team, MDACC Inflammatory Breast Cancer Team and Woodward, Wendy A},
title = {Use of natural language processing to identify patients with inflammatory breast cancer across a health-care system},
journal = {JNCI Cancer Spectrum},
year = {2025},
doi = {10.1093/jncics/pkaf058}
}
EUVIMED