Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: Oct 8, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

676

A cascaded fully convolutional network framework for dilated pancreatic duct segmentation.

Chen Shen1, Holger R Roth2, Yuichiro Hayashi1

  • 1Graduate School of Informatics, Nagoya University, Nagoya, Japan.

International Journal of Computer Assisted Radiology and Surgery
|December 24, 2021
PubMed
Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

A giant retropharyngeal space lipoma: a case report and review of the literature.

Frontiers in surgery·2026
Same author

Acute Kidney Injury Due to Granulomatous Interstitial Nephritis as the Initial Presentation of CLL/SLL Treated with Acalabrutinib.

Internal medicine (Tokyo, Japan)·2026
Same author

Deep learning-based artificial intelligence can improve the diagnosis of small bowel obstruction: stratified comparison study and hierarchical Bayesian model.

Scientific reports·2026
Same author

Blastic plasmacytoid dendritic cell neoplasm following Langerhans cell histiocytosis.

JAAD case reports·2026
Same author

A case study of virtual control group verification using historical control data at a single facility.

The Journal of toxicological sciences·2026
Same author

Tumor-Derived Extracellular Vesicles in Plasma for Predicting Anti-PD-1 Antibody Efficacy in Non-Small Cell Lung Cancer.

Cancer science·2026

This study introduces a novel two-step framework using artificial intelligence to segment dilated pancreatic ducts in CT scans. This automated approach shows potential for early pancreatic cancer screening in individuals without PDAC.

Area of Science:

  • Medical Imaging
  • Artificial Intelligence in Radiology
  • Computational Pathology

Background:

  • Pancreatic duct dilation is an early indicator of pancreatic ductal adenocarcinoma (PDAC).
  • Automated segmentation of dilated pancreatic ducts is challenging due to limited labeled data and significant voxel imbalance.
  • Existing research on dilated pancreatic duct segmentation as a screening tool is scarce.

Purpose of the Study:

  • To propose and evaluate a two-step framework for automated dilated pancreatic duct segmentation from abdominal CT volumes.
  • To address the challenges of limited labeled data and voxel imbalance in segmenting dilated pancreatic ducts.
  • To leverage fully convolutional networks (FCNs) for improved segmentation accuracy.

Main Methods:

  • A cascaded segmentation approach using FCNs was developed.
Keywords:
Computed tomographyPancreatic duct dilationPancreatic duct segmentation

More Related Videos

Modified Single-Loop Reconstruction for Pancreaticoduodenectomy
13:01

Modified Single-Loop Reconstruction for Pancreaticoduodenectomy

Published on: September 28, 2019

7.5K
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

548

Related Experiment Videos

Last Updated: Oct 8, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

676
Modified Single-Loop Reconstruction for Pancreaticoduodenectomy
13:01

Modified Single-Loop Reconstruction for Pancreaticoduodenectomy

Published on: September 28, 2019

7.5K
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

548
  • An initial FCN was trained on a public pancreas dataset to generate a region of interest (ROI) for the pancreas.
  • A 3D U-Net-like FCN with skip connections, an attention mechanism, and a combined Dice and Focal loss function was employed for detailed segmentation.
  • Random data augmentation was utilized to enhance model generalizability.

    • Main Results:

    • A manually curated dataset of dilated pancreatic ducts was created using semi-automated annotation tools.
    • The proposed framework demonstrated practical utility for dilated pancreatic duct segmentation.
    • The average Dice score and sensitivity achieved were 49.9% and 51.9%, respectively.
    • These results indicate the potential of the approach as a clinical screening tool.

    Conclusions:

    • An automated framework for dilated pancreatic duct segmentation was investigated.
    • The cascaded strategy significantly improved pancreatic duct segmentation performance.
    • Modifications to FCNs, combined loss function, and data augmentation facilitated automated segmentation.