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 Concept Videos

You might also read

Related Articles

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

Sort by
Same author

Fibromyxoid Nephrogenic Adenoma Mimicking Renal Cell Carcinoma: A Diagnostic Pitfall in Fine-Needle Aspiration Cytology.

Diagnostic cytopathology·2026
Same author

Real-Time in Vivo Monitoring of Circulating Endometrial Cells: Uncovering Systemic Dissemination and Temporal Fluctuations in Endometriosis.

Advanced science (Weinheim, Baden-Wurttemberg, Germany)·2026
Same author

Frequency-Specific Transcranial Photobiomodulation Elicits Complementary Glial Mechanisms for Neurovascular Protection and Amyloid Clearance in Alzheimer Disease.

Cyborg and bionic systems (Washington, D.C.)·2026
Same author

Circadian rhythm of dendritic cells directs the time-dependent efficacy of photodynamic therapy.

Cell communication and signaling : CCS·2026
Same author

A clinician-centric intelligent method towards reliable pancreatic cancer vascular invasion assessment: a retrospective, multi-centre study.

The Lancet regional health. Western Pacific·2026
Same author

Dc-EEMF: Decision-level constrained multi-focus image fusion for multi-modality photoacoustic microscopy.

IEEE transactions on bio-medical engineering·2026

Related Experiment Video

Updated: Aug 8, 2025

Laparoscopic Pancreatoduodenectomy for Pancreatic Cancer Using In-Situ No-Touch Isolation Technique
08:12

Laparoscopic Pancreatoduodenectomy for Pancreatic Cancer Using In-Situ No-Touch Isolation Technique

Published on: February 2, 2022

2.1K

Generalized pancreatic cancer diagnosis via multiple instance learning and anatomically-guided shape normalization.

Jiaqi Qu1, Xunbin Wei2, Xiaohua Qian1

  • 1School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai 200030, China.

Medical Image Analysis
|February 26, 2023
PubMed
Summary

This study introduces an automated method for early pancreatic cancer diagnosis using contrast-enhanced CT images and multiple instance learning. The approach improves accuracy and generalizability, offering a promising tool for detecting this challenging disease.

Keywords:
Classifier generalizationContrastive learningEarly cancer diagnosisMultiple instance learningPancreatic cancer

More Related Videos

DIPLOMA Approach for Standardized Pathology Assessment of Distal Pancreatectomy Specimens
10:38

DIPLOMA Approach for Standardized Pathology Assessment of Distal Pancreatectomy Specimens

Published on: February 1, 2020

7.5K
Application of End-to-end Anastomosis in Robotic Central Pancreatectomy
10:58

Application of End-to-end Anastomosis in Robotic Central Pancreatectomy

Published on: June 2, 2018

8.1K

Related Experiment Videos

Last Updated: Aug 8, 2025

Laparoscopic Pancreatoduodenectomy for Pancreatic Cancer Using In-Situ No-Touch Isolation Technique
08:12

Laparoscopic Pancreatoduodenectomy for Pancreatic Cancer Using In-Situ No-Touch Isolation Technique

Published on: February 2, 2022

2.1K
DIPLOMA Approach for Standardized Pathology Assessment of Distal Pancreatectomy Specimens
10:38

DIPLOMA Approach for Standardized Pathology Assessment of Distal Pancreatectomy Specimens

Published on: February 1, 2020

7.5K
Application of End-to-end Anastomosis in Robotic Central Pancreatectomy
10:58

Application of End-to-end Anastomosis in Robotic Central Pancreatectomy

Published on: June 2, 2018

8.1K

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Oncology

Background:

  • Pancreatic cancer has a high mortality rate, often diagnosed late due to subtle symptoms.
  • Early detection is crucial for improving patient outcomes but remains a significant clinical challenge.

Purpose of the Study:

  • To develop an automated method for early pancreatic cancer diagnosis using contrast-enhanced CT images.
  • To enhance diagnosis stability and generalizability through novel image processing and machine learning techniques.

Main Methods:

  • Utilized multiple instance learning (MIL) with contrast-enhanced CT images.
  • Implemented anatomically-guided shape normalization for improved feature extraction.
  • Employed instance-level contrastive learning to maintain tumor feature integrity.
  • Applied a balance-adjustment strategy to address class imbalance.

Main Results:

  • Achieved remarkable performance on cross-validation and independent test datasets.
  • Demonstrated significant improvements in diagnostic generalizability.
  • Successfully identified tumors smaller than 2 cm with high accuracy (80.9% and 90.1% in independent tests).

Conclusions:

  • The proposed automated method shows significant potential for the early diagnosis of pancreatic cancer.
  • The integration of shape normalization and contrastive learning enhances diagnostic accuracy and stability.
  • This approach offers a valuable tool for clinical application in pancreatic cancer detection.