EUVIMEDEUVIMED entdecken
VergleichsstudieOpen Access

Evaluation of GPT-5 for Esophageal Cancer Staging Using Fluorodeoxyglucose Positron Emission Tomography Maximum-Intensity Projection Images: Comparative Pilot Study

Hiroki Maruyama iD, Yoshitaka Toyama iD, Yuya Araki iD, Kentaro Takanami iD, Masato Ito iD, Yumi Nakajima iD, Kei Takase iD, Takashi Kamei iD

JMIR Cancer · 2026

DOI: 10.2196/86630 ↗

Kurz gefasst

Vergleichsstudie in der Zeitschrift JMIR Cancer, 2026. Verfasst von Hiroki Maruyama und 7 weiteren Autor:innen. Beteiligte Einrichtung: Department of Surgery, Graduate School of Medicine, Tohoku University, Sendai, Japan. Hauptthemen laut MEDLINE: Esophageal Neoplasms, Fluorodeoxyglucose F18, Positron Emission Tomography Computed Tomography, Positron-Emission Tomography. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 37 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Accurate esophageal cancer staging relies on 18F fluorodeoxyglucose positron emission tomography (18F FDG-PET), but its interpretation is complex and time-intensive. This diagnostic burden is exacerbated by significant workforce shortages in both radiology and surgery, thus necessitating automated support systems. The emergence of advanced large language models (LLMs) has raised expectations for their potential to fulfill this role in complex medical tasks.

Objective

We evaluated the diagnostic accuracy of LLMs for staging esophageal cancer using 18F FDG-PET images, with a focus on their ability to assess lymph nodes (LNs; clinical N [cN]) and distant metastases (clinical M [cM]) for automated radiology reporting.

Methods

This retrospective study included 120 consecutive adult patients who were diagnosed with esophageal squamous cell carcinoma and underwent 18F FDG-PET/computed tomography at Tohoku University Hospital between January 2019 and December 2021. Patients with prior treatment, nonsquamous cell carcinoma histology, or blood glucose levels ≥200 mg/dL were excluded. Frontal maximum-intensity projection positron emission tomography images were extracted, standardized, and analyzed along with information regarding the tumor location. Six LLMs (GPT-5, GPT-4.5, GPT-4.1, OpenAI-o3, -o1, and GPT-4 Turbo) and 4 blinded human evaluators (a nuclear medicine specialist, a gastrointestinal surgeon, and 2 radiology residents) assessed the presence of thoracic and abdominal LN metastases on a region-level basis and determined cN and cM staging on a patient-level basis. The model analyses were performed using the application programming interface in a zero-shot setting. Radiology reports served as the reference standard. Diagnostic agreement and accuracy were evaluated using Cohen κ and the Cochran Q test. Additionally, to account for the class imbalance in the dataset, the Matthews Correlation Coefficient was calculated as a robust metric for binary classification performance. Post hoc McNemar tests were performed with Bonferroni correction; statistical significance for pairwise comparisons was set at P

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Esophageal NeoplasmsFluorodeoxyglucose F18Positron Emission Tomography Computed TomographyPositron-Emission TomographyEsophageal Squamous Cell CarcinomaHumansMaleNeoplasm StagingFemaleRetrospective StudiesLarge Language ModelsMiddle AgedPilot ProjectsAgedRadiopharmaceuticalsLymphatic MetastasisAdult

Stichwörter der Autor:innen

18F FDG-PET imagingLLMsesophageal cancer stagingfluorodeoxyglucose positron emission tomographygenerative artificial intelligencelarge language modelsradiology report automation

Substanzen

Fluorodeoxyglucose F18Radiopharmaceuticals

Autor:innen und Institutionen

  • Department of Surgery, Graduate School of Medicine, Tohoku University, Sendai, Japan.
  • Department of Imaging and Anatomy for Groundbreaking Education Collaborative Research, Graduate School of Medicine, Tohoku University, Sendai, Japan.
  • Department of Diagnostic Radiology, Tohoku University Hospital, Sendai, Japan.
  • School of Medicine, Tohoku University, Sendai, Japan.
  • Department of Diagnostic Radiology, Osaki Citizen Hospital, Osaki, Japan.
  • Department of Diagnostic Radiology, Tohoku Medical and Pharmaceutical University, Sendai, Japan.

ORCID-Profile: Hiroki Maruyama, Yoshitaka Toyama, Yuya Araki, Kentaro Takanami, Masato Ito, Yumi Nakajima, Kei Takase, Takashi Kamei

Zitieren

APA

Maruyama, H., Toyama, Y., Araki, Y., Takanami, K., Ito, M., Nakajima, Y., Takase, K., & Kamei, T. (2026). Evaluation of GPT-5 for Esophageal Cancer Staging Using Fluorodeoxyglucose Positron Emission Tomography Maximum-Intensity Projection Images: Comparative Pilot Study. JMIR Cancer. https://doi.org/10.2196/86630

Vancouver

Maruyama H, Toyama Y, Araki Y, Takanami K, Ito M, Nakajima Y, et al. Evaluation of GPT-5 for Esophageal Cancer Staging Using Fluorodeoxyglucose Positron Emission Tomography Maximum-Intensity Projection Images: Comparative Pilot Study. JMIR Cancer. 2026. doi:10.2196/86630

BibTeX
@article{Maruyama2026,
  author = {Maruyama, Hiroki and Toyama, Yoshitaka and Araki, Yuya and Takanami, Kentaro and Ito, Masato and Nakajima, Yumi and Takase, Kei and Kamei, Takashi},
  title = {Evaluation of GPT-5 for Esophageal Cancer Staging Using Fluorodeoxyglucose Positron Emission Tomography Maximum-Intensity Projection Images: Comparative Pilot Study},
  journal = {JMIR Cancer},
  year = {2026},
  doi = {10.2196/86630}
}
Kostenlos starten →