EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Differentiating localized autoimmune pancreatitis and pancreatic ductal adenocarcinoma using endoscopic ultrasound images with deep learning

Hitomi Nakamura iD, Motohisa Fukuda, Akiko Matsuda, Naohiko Makino, Hirohito Kimura, Yu Ohtaki, Yoshihito Nawa, Soushi Oyama, Yuya Suzuki, Toshikazu Kobayashi, Tetsuya Ishizawa, Yasuharu Kakizaki und 1 weitere

DEN Open · 2024-04 · 4 (1)

DOI: 10.1002/deo2.344 ↗

Kurz gefasst

Fachartikel in der Zeitschrift DEN Open, online erschienen am 02.03.2024. Verfasst von Hitomi Nakamura und 12 weiteren Autor:innen. Beteiligte Einrichtung: Department of Gastroenterology Faculty of Medicine Yamagata University Yamagata Japan. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 24 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Objectives

Localized autoimmune pancreatitis is difficult to differentiate from pancreatic ductal adenocarcinoma on endoscopic ultrasound images. In recent years, deep learning methods have improved the diagnosis of diseases. Hence, we developed a special cross‐validation framework to search for effective methodologies of deep learning in distinguishing autoimmune pancreatitis from pancreatic ductal adenocarcinoma on endoscopic ultrasound images.

Methods

Data from 24 patients diagnosed with localized autoimmune pancreatitis (8751 images) and 61 patients diagnosed with pancreatic ductal adenocarcinoma (20,584 images) were collected from 2016 to 2022. We applied transfer learning to a convolutional neural network called ResNet152, together with our innovative imaging method contributing to data augmentation and temporal data process. We divided patients into five groups according to different factors for 5‐fold cross‐validation, where the ordered and balanced datasets were created for the performance evaluations.

Results

ResNet152 surpassed the endoscopists in all evaluation metrics with almost all datasets. Interestingly, when the dataset is balanced according to the factor of the endoscopists’ diagnostic accuracy, the area under the receiver operating characteristic curve and accuracy were highest at 0.85 and 0.80, respectively.

Conclusions

It is deduced that image features useful for ResNet152 correlate with those used by endoscopists for their diagnoses. This finding may contribute to sample‐efficient dataset preparation to train convolutional neural networks for endoscopic ultrasonography‐imaging diagnosis.

Schlagwörter

Stichwörter der Autor:innen

deep learningdiagnostic imagingendoscopic ultrasound imageslocalized autoimmune pancreatitispancreatic ductal adenocarcinoma

Autor:innen und Institutionen

  • Department of Gastroenterology Faculty of Medicine Yamagata University Yamagata Japan.
  • Department of Science Faculty of Science Yamagata University Yamagata Japan.
  • Institute of Well-Being Yamagata University Yamagata Japan.

ORCID-Profile: Hitomi Nakamura

Zitieren

APA

Nakamura, H., Fukuda, M., Matsuda, A., Makino, N., Kimura, H., Ohtaki, Y., Nawa, Y., Oyama, S., Suzuki, Y., Kobayashi, T., Ishizawa, T., Kakizaki, Y., & Ueno, Y. (2024). Differentiating localized autoimmune pancreatitis and pancreatic ductal adenocarcinoma using endoscopic ultrasound images with deep learning. DEN Open, 4(1). https://doi.org/10.1002/deo2.344

Vancouver

Nakamura H, Fukuda M, Matsuda A, Makino N, Kimura H, Ohtaki Y, et al. Differentiating localized autoimmune pancreatitis and pancreatic ductal adenocarcinoma using endoscopic ultrasound images with deep learning. DEN Open. 2024;4(1). doi:10.1002/deo2.344

BibTeX
@article{Nakamura2024,
  author = {Nakamura, Hitomi and Fukuda, Motohisa and Matsuda, Akiko and Makino, Naohiko and Kimura, Hirohito and Ohtaki, Yu and Nawa, Yoshihito and Oyama, Soushi and Suzuki, Yuya and Kobayashi, Toshikazu and Ishizawa, Tetsuya and Kakizaki, Yasuharu and Ueno, Yoshiyuki},
  title = {Differentiating localized autoimmune pancreatitis and pancreatic ductal adenocarcinoma using endoscopic ultrasound images with deep learning},
  journal = {DEN Open},
  year = {2024},
  volume = {4},
  number = {1},
  doi = {10.1002/deo2.344}
}
Kostenlos starten →