EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Large language models for pneumonia detection in radiology reports via text analysis

Weisheng Chen, Xinya Li, Zhongkai Qu, Ting Liu, Bo Chen, Liying Huang, Ningxia He, Zhigang Wang, Jun Lyu iD

PLOS Digital Health · 23.09.2026 · 5 (9) · e0001104

DOI: 10.1371/journal.pdig.0001104 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS Digital Health, online erschienen am 23.09.2026. Verfasst von Weisheng Chen und 8 weiteren Autor:innen. Beteiligte Einrichtung: Department of Intensive Care Unit, The First Affiliated Hospital of Jinan University, Guangzhou, China. Die Arbeit nennt 37 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Guangzhou Municipal Science and Technology Program key projects, National Natural Science Foundation of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Pneumonia is a common infection in critically ill patients with poor outcomes, and its identification relies heavily on radiological evidence. Most MIMIC-based pneumonia studies rely on structured codes rather than radiological text, which may limit the precision of pneumonia-related case identification. This study evaluated large language models (LLMs) and retrieval-augmented generation (RAG) frameworks for automated detection of radiographic evidence of pneumonia from radiology reports. Radiological reports from the MIMIC-IV database were analyzed using pre-trained models in an inference-only setting without additional training, fine-tuning, or dataset splitting. Model performance was evaluated using accuracy, precision, sensitivity, specificity, and F1-score, with point estimates calculated from the full dataset and corresponding 95% confidence intervals estimated using 1,000 patient-level clustered bootstrap replicates. Independent evaluation was subsequently performed using a GZ dataset from a tertiary hospital in Guangzhou. RAG generally improved model performance, with gemma3-27b achieving the highest accuracy of 96.4% on the MIMIC-IV dataset and maintaining an accuracy of 87.6% on the GZ dataset. Model performance did not increase linearly with parameter size, suggesting that task-specific model suitability may be more important than parameter scale for this application. Although larger models required longer inference times, their increased computational costs did not consistently translate into proportional performance gains. Overall, LLMs demonstrated high accuracy in interpreting radiology reports for detecting radiographic evidence of pneumonia. The RAG architecture further enhanced robustness and generalizability through knowledge augmentation, providing a methodological basis for automated radiology report analysis and large-scale database curation.

Autor:innen und Institutionen

  • Department of Intensive Care Unit, The First Affiliated Hospital of Jinan University, Guangzhou, China.
  • School of Nursing, Jinan University, Guangzhou, China.
  • Department of Critical Care Medicine, Shenzhen People's Hospital, The Second Clinical Medical College, Jinan University, Shenzhen, China.
  • Department of Pharmacy, The First Affiliated Hospital of Jinan University, Guangzhou, China.
  • Clinical Research Center, The First Affiliated Hospital of Jinan University, Guangzhou, China.

ORCID-Profile: Jun Lyu

Förderung

  • Guangzhou Municipal Science and Technology Program key projects
  • National Natural Science Foundation of China

Zitieren

APA

Chen, W., Li, X., Qu, Z., Liu, T., Chen, B., Huang, L., He, N., Wang, Z., & Lyu, J. (2026). Large language models for pneumonia detection in radiology reports via text analysis. PLOS Digital Health, 5(9), e0001104. https://doi.org/10.1371/journal.pdig.0001104

Vancouver

Chen W, Li X, Qu Z, Liu T, Chen B, Huang L, et al. Large language models for pneumonia detection in radiology reports via text analysis. PLOS Digital Health. 2026;5(9):e0001104. doi:10.1371/journal.pdig.0001104

BibTeX
@article{Chen2026,
  author = {Chen, Weisheng and Li, Xinya and Qu, Zhongkai and Liu, Ting and Chen, Bo and Huang, Liying and He, Ningxia and Wang, Zhigang and Lyu, Jun},
  title = {Large language models for pneumonia detection in radiology reports via text analysis},
  journal = {PLOS Digital Health},
  year = {2026},
  volume = {5},
  number = {9},
  pages = {e0001104},
  doi = {10.1371/journal.pdig.0001104}
}
Kostenlos starten →