EUVIMEDScopri EUVIMED
VergleichsstudieOpen access

Large Language Model Versus Manual Review for Clinical Data Curation in Breast Cancer: Retrospective Comparative Study

Young-Joon Kang iD, Hocheol Lee iD, Jae Pak Yi iD, Hyobin Kim iD, Chang Ik Yoon iD, Jong Min Baek iD, Yong-Seok Kim iD, Ye Won Jeon iD, Jiyoung Rhu iD, Su Hyun Lim iD, Hoon Choi iD, Se Jeong Oh iD

JMIR Medical Informatics · 2025-11-06 · 13 · e73605-e73605

DOI: 10.2196/73605 ↗

At a glance

Vergleichsstudie in JMIR Medical Informatics, published online on 2025-11-06. Written by Young-Joon Kang and 11 co-authors. Institution: Department of Surgery, College of Medicine, The Catholic University of Korea, Incheon St Mary's Hospital, 56, Dongsu-ro, Bupyeong-gu, Incheon, 21431,…. Major MEDLINE topics: Breast Neoplasms, Electronic Health Records, Data Curation. Cited 13 times so far (Crossref); 23 references. The full text is freely available in PubMed Central.

AbstractSubjectsDati bibliograficiCiteRelated studies

Abstract (original language)

Background

Manual review of electronic health records for clinical research is labor-intensive and prone to reviewer-dependent variations. Large language models (LLMs) offer potential for automated clinical data extraction; however, their feasibility in surgical oncology remains underexplored.

Objective

This study aimed to evaluate the feasibility and accuracy of LLM-based processing compared with manual physician review for extracting clinical data from breast cancer records.

Methods

We conducted a retrospective comparative study analyzing breast cancer records from 5 academic hospitals (January 2019-December 2019). Two data extraction pathways were compared: (1) manual physician review with direct electronic health record access (group 1: 1366/3100, 44.06%) and (2) LLM-based processing using Claude 3.5 Sonnet (Anthropic) on deidentified data automatically extracted through a clinical data warehouse platform (group 2: 1734/3100, 55.94%). The automated extraction system provided prestructured, deidentified data sheets organized by clinical domains, which were then processed by the LLM. The LLM prompt was developed through a 3-phase iterative process over 2 days. Primary outcomes included missing value rates, extraction accuracy, and concordance between groups. Secondary outcomes included comparison with the Korean Breast Cancer Society national registry data, processing time, and resource use. Validation involved 50 stratified random samples per group (900 data points each), assessed by 4 breast surgical oncologists. Statistical analysis included chi-square tests, 2-tailed t tests, Cohen κ, and intraclass correlation coefficients. The accuracy threshold was set at 90%.

Results

The LLM achieved 90.8% (817) accuracy in validation analysis. Missing data patterns differed between groups: group 2 showed better lymph node documentation (missing: 152/1734, % vs 294/1366, %) but higher missing rates for cancer staging (211/1734, 12.17% vs 43/1366, 3.15%). Both groups demonstrated similar breast-conserving surgery rates (1107/1734, 84% vs 868/1366, %). Processing efficiency differed substantially: LLM processing required 12 days with 2 physicians versus 7 months with 5 physicians for manual review, representing a 91% reduction in physician hours (96 h vs 1025 h). The LLM group captured significantly more survival events (41 vs 11; P =.002). Stage distribution in the LLM group aligned better with national registry data (Cramér V=0.03 vs 0.07). Application programming interface costs totaled US $260 for 1734 cases (US $0.15 per case).

Conclusions

LLM-based curation of automatically extracted, deidentified clinical data demonstrated comparable effectiveness to manual physician review while reducing processing time by 95% and physician hours by 91%. This 2-step approach—automated data extraction followed by LLM curation—addresses both privacy concerns and efficiency needs. Despite limitations in integrating multiple clinical events, this methodology offers a scalable solution for clinical data extraction in oncology research. The 90.8% accuracy rate and superior capture of survival events suggest that combining automated data extraction systems with LLM processing can accelerate retrospective clinical research while maintaining data quality and patient privacy.

Subjects

MeSH terms (MEDLINE)

Breast NeoplasmsElectronic Health RecordsData CurationHumansRetrospective StudiesFemaleNatural Language ProcessingLarge Language Models

Author keywords

artificial intelligencebreast neoplasmsclinical oncologydata mininglarge language modelnatural language processing

Authors and institutions

  • Department of Surgery, College of Medicine, The Catholic University of Korea, Incheon St Mary's Hospital, 56, Dongsu-ro, Bupyeong-gu, Incheon, 21431, Republic of Korea, 01026383847.
  • Department of AI Health Information Management, Yonsei University (Mirae), Wonju, Republic of Korea.
  • Department of Surgery, College of Medicine, The Catholic University of Korea, Seoul St Mary's Hospital, Seoul, Republic of Korea.
  • Department of Surgery, College of Medicine, The Catholic University of Korea, Yeouido St Mary's Hospital, Seoul, Republic of Korea.
  • Department of Surgery, College of Medicine, The Catholic University of Korea, Uijeongbu St Mary's Hospital, Uijeongbu, Republic of Korea.
  • Department of Surgery, College of Medicine, The Catholic University of Korea, St Vincent's Hospital, Suwon, Republic of Korea.

ORCID profiles: Young-Joon Kang, Hocheol Lee, Jae Pak Yi, Hyobin Kim, Chang Ik Yoon, Jong Min Baek, Yong-Seok Kim, Ye Won Jeon, Jiyoung Rhu, Su Hyun Lim, Hoon Choi, Se Jeong Oh

Cite

APA

Kang, Y. J., Lee, H., Yi, J. P., Kim, H., Yoon, C. I., Baek, J. M., Kim, Y. S., Jeon, Y. W., Rhu, J., Lim, S. H., Choi, H., & Oh, S. J. (2025). Large Language Model Versus Manual Review for Clinical Data Curation in Breast Cancer: Retrospective Comparative Study. JMIR Medical Informatics, 13, e73605-e73605. https://doi.org/10.2196/73605

Vancouver

Kang YJ, Lee H, Yi JP, Kim H, Yoon CI, Baek JM, et al. Large Language Model Versus Manual Review for Clinical Data Curation in Breast Cancer: Retrospective Comparative Study. JMIR Medical Informatics. 2025;13:e73605-e73605. doi:10.2196/73605

BibTeX
@article{Kang2025,
  author = {Kang, Young-Joon and Lee, Hocheol and Yi, Jae Pak and Kim, Hyobin and Yoon, Chang Ik and Baek, Jong Min and Kim, Yong-Seok and Jeon, Ye Won and Rhu, Jiyoung and Lim, Su Hyun and Choi, Hoon and Oh, Se Jeong},
  title = {Large Language Model Versus Manual Review for Clinical Data Curation in Breast Cancer: Retrospective Comparative Study},
  journal = {JMIR Medical Informatics},
  year = {2025},
  volume = {13},
  pages = {e73605--e73605},
  doi = {10.2196/73605}
}
Start for free →