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

Steps in Outbreak Investigation01:18

Steps in Outbreak Investigation

In the ever-evolving field of public health, statistical analysis serves as a cornerstone for understanding and managing disease outbreaks. By leveraging various statistical tools, health professionals can predict potential outbreaks, analyze ongoing situations, and devise effective responses to mitigate impact. For that to happen, there are a few possible stages of the analysis:
Infectious Diseases and Their Occurrence01:28

Infectious Diseases and Their Occurrence

Infectious diseases appear in populations through various transmission patterns, influenced by pathogen characteristics, population immunity, environmental conditions, and social behavior. Understanding these patterns is essential for effective public health surveillance and intervention. These categories—sporadic, outbreak, epidemic, pandemic, and endemic—help frame the nature and scope of disease events.Sporadic diseases occur irregularly and infrequently, without a predictable temporal or...
Rabies01:28

Rabies

Rabies is a lethal zoonotic disease caused by a single-stranded, negative-sense RNA virus of the Lyssavirus genus, within the family Rhabdoviridae. Its primary mode of transmission to humans is through bites or saliva-contaminated scratches from infected mammals such as dogs, bats, raccoons, or foxes. Transmission can also occur if infectious saliva contacts abraded skin or intact mucous membranes, including the conjunctiva.Viral Entry and Early ReplicationOnce introduced at the bite or scratch...

You might also read

Related Articles

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

Sort by
Same author

Neuro-symbolic artificial intelligence in medicine.

Nature biomedical engineering·2026
Same author

Correction: SIRT2 Inhibition promotes microglia LC3-associated phagocytosis via NRF2/CD36 after the experimental subarachnoid hemorrhage.

Journal of neuroinflammation·2026
Same author

Identification and Experimental Verification of Key Genes Related to the Ras Signaling Pathway and the Hippo Signaling Pathway in Osteoarthritis Based on Transcriptome Data.

The journal of gene medicine·2026
Same author

Visible-Light-Induced Radical Cascade Sulfonylation/Cyclization to Access Sulfonylated Benzazepines and Benzoxepines by EDA Complexes.

The Journal of organic chemistry·2026
Same author

The role of Poller screws in intramedullary nailing for lower limb extra-isthmic fractures.

Journal of orthopaedic surgery and research·2026
Same author

Dataset Distillation via a Noise-Unconstrained Generative Model.

IEEE transactions on pattern analysis and machine intelligence·2026

Related Experiment Video

Updated: May 30, 2026

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

2.7K

SSM-Net: Semi-supervised multi-task network for joint lesion segmentation and classification from pancreatic EUS

Jiajia Li1, Pingping Zhang2, Xia Yang3

  • 1School of Chemistry and Chemical Engineering and National Center for Translational Medicine, Shanghai Jiao Tong University, Shanghai, China.

Artificial Intelligence in Medicine
|June 28, 2024
PubMed
Summary

This study introduces a novel semi-supervised multi-task network (SSM-Net) for improved pancreatic cancer detection. The method enhances early diagnosis by accurately classifying and segmenting lesions in endoscopic ultrasonography (EUS) images, outperforming existing techniques.

Keywords:
Contrastive learningEndoscopic ultrasonographyMulti-task learningNeural networkPancreatic diseasesSemi-supervised learning

More Related Videos

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application
05:56

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application

Published on: April 14, 2023

2.4K
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

385

Related Experiment Videos

Last Updated: May 30, 2026

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

2.7K
Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application
05:56

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application

Published on: April 14, 2023

2.4K
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

385

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Oncology

Background:

  • Pancreatic cancer diagnosis is challenging due to non-specific symptoms and difficulties in early detection using current imaging methods.
  • Endoscopic ultrasonography (EUS) is crucial for pancreatic disease diagnosis, but B-mode imaging is susceptible to artifacts, complicating lesion analysis.
  • Accurate segmentation and classification of pancreatic lesions are essential but require specialized expertise.

Purpose of the Study:

  • To develop a semi-supervised multi-task network (SSM-Net) for joint classification and segmentation of pancreatic lesions in EUS images.
  • To leverage both labeled and unlabeled EUS data to improve diagnostic accuracy and reduce the burden on specialists.
  • To enhance the early detection and characterization of pancreatic cancer through advanced AI techniques.

Main Methods:

  • A saliency-aware representation learning module (SRLM) was developed using unlabeled EUS images to train a feature extraction encoder.
  • A spectral residual module (SRM) generates semantic saliency maps for contrastive loss computation.
  • Channel attention blocks (CABs), a merged global attention module (MGAM), and feature similarity loss (FSL) were employed for lesion segmentation and classification on labeled data.

Main Results:

  • The proposed SSM-Net demonstrated superior performance in pancreatic lesion classification and segmentation compared to state-of-the-art methods.
  • Experiments were conducted on a large-scale EUS-based pancreas image dataset (LS-EUSPI) and a public thyroid gland dataset.
  • The model effectively utilized both labeled and unlabeled data, showcasing the power of semi-supervised learning in medical image analysis.

Conclusions:

  • The SSM-Net offers a promising approach for improving the accuracy and efficiency of pancreatic lesion detection in EUS imaging.
  • This AI-driven method has the potential to aid clinicians in earlier and more precise diagnosis of pancreatic cancer.
  • The study highlights the effectiveness of integrating representation learning and attention mechanisms in semi-supervised medical image analysis.