EUVIMEDEUVIMED entdecken
Open Access

Personalized Impression Generation for PET Reports Using Large Language Models

Xin Tie, Muheon Shin, Ali Pirasteh, Nevein Ibrahim, Zachary Huemann, Sharon M Castellino, Kara M Kelly, John Garrett, Junjie Hu, Steve Y Cho, Tyler J Bradshaw iD

Journal of Imaging Informatics in Medicine · 02.02.2024 · 37 (2) · 471-488

DOI: 10.1007/s10278-024-00985-3 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Imaging Informatics in Medicine, online erschienen am 02.02.2024. Verfasst von Xin Tie und 10 weiteren Autor:innen. Beteiligte Einrichtung: Department of Radiology, School of Medicine and Public Health, University of Wissconsin, Madison, WI, USA. Hauptthemen laut MEDLINE: Positron-Emission Tomography, Natural Language Processing. Die Arbeit wurde bisher 35-mal zitiert (Crossref) und nennt 49 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: NIBIB NIH HHS, NCI NIH HHS, Imaging and Radiology Oncology Core Rhode Island.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Large language models (LLMs) have shown promise in accelerating radiology reporting by summarizing clinical findings into impressions. However, automatic impression generation for whole-body PET reports presents unique challenges and has received little attention. Our study aimed to evaluate whether LLMs can create clinically useful impressions for PET reporting. To this end, we fine-tuned twelve open-source language models on a corpus of 37,370 retrospective PET reports collected from our institution. All models were trained using the teacher-forcing algorithm, with the report findings and patient information as input and the original clinical impressions as reference. An extra input token encoded the reading physician's identity, allowing models to learn physician-specific reporting styles. To compare the performances of different models, we computed various automatic evaluation metrics and benchmarked them against physician preferences, ultimately selecting PEGASUS as the top LLM. To evaluate its clinical utility, three nuclear medicine physicians assessed the PEGASUS-generated impressions and original clinical impressions across 6 quality dimensions (3-point scales) and an overall utility score (5-point scale). Each physician reviewed 12 of their own reports and 12 reports from other physicians. When physicians assessed LLM impressions generated in their own style, 89% were considered clinically acceptable, with a mean utility score of 4.08/5. On average, physicians rated these personalized impressions as comparable in overall utility to the impressions dictated by other physicians (4.03, P = 0.41). In summary, our study demonstrated that personalized impressions generated by PEGASUS were clinically useful in most cases, highlighting its potential to expedite PET reporting by automatically drafting impressions.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Positron-Emission TomographyNatural Language ProcessingHumansRetrospective StudiesAlgorithmsLarge Language Models

Stichwörter der Autor:innen

InformaticsLarge Language ModelsNatural Language ProcessingNuclear MedicinePositron Emission TomographyRadiology Report Summarization

Autor:innen und Institutionen

  • Department of Radiology, School of Medicine and Public Health, University of Wissconsin, Madison, WI, USA.
  • Department of Medical Physics, School of Medicine and Public Health, University of Wisconsin, Madison, WI, USA.
  • Department of Pediatrics, Emory University School of Medicine, Atlanta, GA, USA.
  • Aflac Cancer and Blood Disorders Center, Childrens Healthcare of Atlanta, Atlanta, GA, USA.
  • Department of Pediatric Oncology, Roswell Park Comprehensive Cancer Center, Buffalo, NY, USA.
  • Department of Pediatrics, University at Buffalo Jacobs School of Medicine and Biomedical Sciences, Buffalo, NY, USA.

ORCID-Profile: Tyler J Bradshaw

Förderung

  • NIBIB NIH HHS
  • NCI NIH HHS
  • Imaging and Radiology Oncology Core Rhode Island
  • Biomarker, Imaging and Quality of Life Studies Funding Program
  • NCTN Operations Center Grant
  • NCTN Statistics and Data Center Grant
  • St. Baldrick’s Foundation
  • National Institute of Biomedical Imaging and Bioengineering

Zitieren

APA

Tie, X., Shin, M., Pirasteh, A., Ibrahim, N., Huemann, Z., Castellino, S. M., Kelly, K. M., Garrett, J., Hu, J., Cho, S. Y., & Bradshaw, T. J. (2024). Personalized Impression Generation for PET Reports Using Large Language Models. Journal of Imaging Informatics in Medicine, 37(2), 471-488. https://doi.org/10.1007/s10278-024-00985-3

Vancouver

Tie X, Shin M, Pirasteh A, Ibrahim N, Huemann Z, Castellino SM, et al. Personalized Impression Generation for PET Reports Using Large Language Models. Journal of Imaging Informatics in Medicine. 2024;37(2):471-488. doi:10.1007/s10278-024-00985-3

BibTeX
@article{Tie2024,
  author = {Tie, Xin and Shin, Muheon and Pirasteh, Ali and Ibrahim, Nevein and Huemann, Zachary and Castellino, Sharon M and Kelly, Kara M and Garrett, John and Hu, Junjie and Cho, Steve Y and Bradshaw, Tyler J},
  title = {Personalized Impression Generation for PET Reports Using Large Language Models},
  journal = {Journal of Imaging Informatics in Medicine},
  year = {2024},
  volume = {37},
  number = {2},
  pages = {471--488},
  doi = {10.1007/s10278-024-00985-3}
}
Kostenlos starten →