EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Utility of multimodal deep learning model to diagnose lymph node metastasis in esophageal cancer using computed tomography and positron emission tomography images

Yasuharu Shinozaki, Hirotaka Ishida, Tomofumi Kaneno, Eichi Takaya, Tomoya Kobayashi, Yusuke Taniyama, Chiaki Sato, Hiroshi Okamoto, Yohei Ozawa, Takuya Ueda, Takashi Kamei

Surgery Today · 2025

DOI: 10.1007/s00595-025-03152-5 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Surgery Today, 2025. Verfasst von Yasuharu Shinozaki und 10 weiteren Autor:innen. Beteiligte Einrichtung: Department of Surgery, Tohoku University Graduate School of Medicine, 1-1 Seiryo-Machi, Aoba-Ku, Sendai, Miyagi, 980-8574, Japan. Hauptthemen laut MEDLINE: Esophageal Neoplasms, Deep Learning, Lymphatic Metastasis, Tomography, X-Ray Computed. Die Arbeit nennt 39 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: CREST Grant, JSPS KAKENHI, MEXT/JSPS WISE Program: Advanced Graduate Program for Future Medicine and Health Care, Tohoku University.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Purpose

This study aimed to assess the performance of a deep learning model using multimodal imaging for detecting lymph node metastasis in esophageal cancer in comparison to expert assessments.

Methods

A retrospective analysis was performed for 521 lymph nodes from 167 patients with esophageal cancer who underwent esophagectomy. Deep learning models were developed based on multimodal imaging, including non-contrast-enhanced computed tomography, contrast-enhanced computed tomography, and positron emission tomography imaging. The diagnostic performance was evaluated and compared with expert assessments using a receiver operating characteristic curve analysis.

Results

The area under the receiver operating characteristic curve values for the deep learning model were 0.81 with multimodal imaging, 0.73 with non-contrast-enhanced computed tomography, 0.72 with contrast-enhanced computed tomography, and 0.75 with positron emission tomography were calculated. The area under the curve of the deep learning model (0.81) demonstrated diagnostic performance comparable to that of experienced experts (area under the curve, 0.84; P = 0.62, DeLong’s test).

Conclusion

The multimodal deep learning model using computed tomography and positron emission tomography demonstrated performance comparable to that of experts in diagnosing the presence of lymph node metastasis, a key prognostic factor in esophageal cancer, suggesting its potential clinical utility.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Esophageal NeoplasmsDeep LearningLymphatic MetastasisTomography, X-Ray ComputedPositron-Emission TomographyMultimodal ImagingLymph NodesHumansRetrospective StudiesMaleFemaleMiddle AgedAgedEsophagectomyROC CurveAged, 80 and over

Stichwörter der Autor:innen

Computed tomographyEsophageal cancerLymph node metastasisMultimodal deep learning modelPositron emission tomography

Autor:innen und Institutionen

  • Department of Surgery, Tohoku University Graduate School of Medicine, 1-1 Seiryo-Machi, Aoba-Ku, Sendai, Miyagi, 980-8574, Japan.
  • Department of Surgery, Tohoku University Graduate School of Medicine, 1-1 Seiryo-Machi, Aoba-Ku, Sendai, Miyagi, 980-8574, Japan. h-ishida@tohoku.ac.jp.
  • Department of Diagnostic Imaging, Tohoku University Graduate School of Medicine, Sendai, Japan.
  • AI Lab, Tohoku University Hospital, Sendai, Japan.
  • Department of Diagnostic Radiology, Tohoku University Graduate School of Medicine, Sendai, Japan.

Förderung

  • CREST Grant
  • JSPS KAKENHI
  • MEXT/JSPS WISE Program: Advanced Graduate Program for Future Medicine and Health Care, Tohoku University

Zitieren

APA

Shinozaki, Y., Ishida, H., Kaneno, T., Takaya, E., Kobayashi, T., Taniyama, Y., Sato, C., Okamoto, H., Ozawa, Y., Ueda, T., & Kamei, T. (2025). Utility of multimodal deep learning model to diagnose lymph node metastasis in esophageal cancer using computed tomography and positron emission tomography images. Surgery Today. https://doi.org/10.1007/s00595-025-03152-5

Vancouver

Shinozaki Y, Ishida H, Kaneno T, Takaya E, Kobayashi T, Taniyama Y, et al. Utility of multimodal deep learning model to diagnose lymph node metastasis in esophageal cancer using computed tomography and positron emission tomography images. Surgery Today. 2025. doi:10.1007/s00595-025-03152-5

BibTeX
@article{Shinozaki2025,
  author = {Shinozaki, Yasuharu and Ishida, Hirotaka and Kaneno, Tomofumi and Takaya, Eichi and Kobayashi, Tomoya and Taniyama, Yusuke and Sato, Chiaki and Okamoto, Hiroshi and Ozawa, Yohei and Ueda, Takuya and Kamei, Takashi},
  title = {Utility of multimodal deep learning model to diagnose lymph node metastasis in esophageal cancer using computed tomography and positron emission tomography images},
  journal = {Surgery Today},
  year = {2025},
  doi = {10.1007/s00595-025-03152-5}
}
Kostenlos starten →