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

[Fluid management in orthotopic liver transplantation].

Zhongguo wei zhong bing ji jiu yi xue = Chinese critical care medicine = Zhongguo weizhongbing jijiuyixue·2006
Same author

Electromagnetic modelling of Raman enhancement from nanoscale substrates: a route to estimation of the magnitude of the chemical enhancement mechanism in SERS.

Faraday discussions·2006
Same author

[Experimental study on protective effects of HupA in the treatment of isocarbophos poisoning].

Zhonghua lao dong wei sheng zhi ye bing za zhi = Zhonghua laodong weisheng zhiyebing zazhi = Chinese journal of industrial hygiene and occupational diseases·2006
Same author

[Complete sequence and gene organization of the Tibetan chicken mitochondrial genome].

Yi chuan = Hereditas·2006
Same author

Liver microcirculation after hepatic artery embolization with degradable starch microspheres in vivo.

World journal of gastroenterology·2006
Same author

A recyclable fluorous (S)-pyrrolidine sulfonamide promoted direct, highly enantioselective Michael addition of ketones and aldehydes to nitroolefins in water.

Organic letters·2006

Related Experiment Video

Updated: Jul 17, 2025

Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images
08:20

Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images

Published on: October 27, 2023

1.5K

Hybrid-supervised bidirectional transfer networks for computer-aided diagnosis.

Ronglin Gong1, Jing Shi2, Jian Wang1

  • 1Key Laboratory of Specialty Fiber Optics and Optical Access Networks, Joint International Research Laboratory of Specialty Fiber Optics and Advanced Communication, School of Communication and Information Engineering, Shanghai University, China; Shanghai Institute for Advanced Communication and Data Science, Shanghai University, China.

Computers in Biology and Medicine
|September 6, 2023
PubMed
Summary

This study introduces Hybrid-supervised Bidirectional Transfer Networks (HBTN), a novel self-supervised learning framework that enhances computer-aided diagnosis (CAD) models. HBTN improves diagnostic accuracy, especially with limited medical imaging data.

Keywords:
Bidirectional transferComputer-aided diagnosisGray-scale image mappingHybrid-supervised learningSelf-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

2.5K
A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
04:23

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

Published on: April 21, 2023

1.9K

Related Experiment Videos

Last Updated: Jul 17, 2025

Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images
08:20

Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images

Published on: October 27, 2023

1.5K
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

2.5K
A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
04:23

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images

Published on: April 21, 2023

1.9K

Area of Science:

  • Medical Imaging
  • Artificial Intelligence
  • Computer-Aided Diagnosis

Background:

  • Medical imaging is crucial for disease diagnosis, but accuracy relies on radiologist skill.
  • Computer-aided diagnosis (CAD) aims to improve diagnostic accuracy, consistency, and reproducibility.
  • Convolutional Neural Networks (CNNs) show promise in CAD but struggle with small datasets; Self-supervised learning (SSL) is effective for limited medical imaging data, yet conventional frameworks may not optimally pre-train downstream tasks.

Purpose of the Study:

  • To propose an improved self-supervised learning (SSL) framework, Hybrid-supervised Bidirectional Transfer Networks (HBTN), to enhance the performance of computer-aided diagnosis (CAD) models, particularly when dealing with limited training samples.
  • To introduce a novel Gray-Scale Image Mapping (GSIM) pretext task that embeds class label information into image restoration for improved discriminative feature learning.
  • To integrate image restoration and classification networks within a unified hybrid-supervised learning (HSL) framework for collaborative knowledge transfer.

Main Methods:

  • Developed Hybrid-supervised Bidirectional Transfer Networks (HBTN), an improved SSL framework for medical image analysis.
  • Introduced a novel Gray-Scale Image Mapping (GSIM) pretext task, enhancing traditional image restoration by incorporating class label information.
  • Integrated image restoration and classification networks into a unified hybrid-supervised learning (HSL) framework for joint training and knowledge transfer.

Main Results:

  • The proposed HBTN framework demonstrated superior performance compared to conventional SSL algorithms in CAD tasks with limited training samples.
  • Experimental results on two medical image datasets validated the effectiveness of HBTN in improving CAD model performance.
  • The GSIM task and HSL framework within HBTN effectively improved discriminative feature learning and downstream task performance.

Conclusions:

  • HBTN offers a significant advancement in self-supervised learning for medical image analysis, addressing the challenge of small sample sizes in CAD.
  • The novel GSIM task and the HSL framework contribute to more effective feature learning and knowledge transfer, boosting CAD model accuracy.
  • HBTN shows strong potential for improving diagnostic accuracy and reliability in medical imaging applications with limited data.