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

Ambient AI in primary care: an exploratory mixed methods survey of UK general practitioners.

BMJ health & care informatics·2026
Same author

Heat risk communication for older adults living in the community: A scoping review.

The journal of climate change and health·2026
Same author

Flexible and scalable federated learning with deep feature prompts for digital pathology.

NPJ digital medicine·2026
Same author

Behavioural nudges to reduce low-value care for low back pain in the emergency department (NUDG-ED): a 2 × 2 factorial, pragmatic cluster randomized trial.

CMAJ : Canadian Medical Association journal = journal de l'Association medicale canadienne·2026
Same author

AI Scribes: Are We Measuring What Matters?

JMIR medical informatics·2026
Same author

Deep Learning Framework for Classifying Whole-Slide Multiplex Immunofluorescence Images to Predict Immunotherapy Response in Melanoma Patients.

IEEE journal of biomedical and health informatics·2025

Related Experiment Video

Updated: Sep 28, 2025

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

3.0K

Weak label based Bayesian U-Net for optic disc segmentation in fundus images.

Hao Xiong1, Sidong Liu1, Roneel V Sharan1

  • 1Centre for Health Informatics, Australian Institute of Health Innovation, Macquarie University, Sydney, Australia.

Artificial Intelligence in Medicine
|March 29, 2022
PubMed
Summary

This study introduces a novel Bayesian U-Net method for optic disc segmentation in fundus images using weak labels. This approach reduces manual annotation time and variance in ophthalmic disease analysis.

Keywords:
Bayesian U-NetExpectation-maximizationFundus imageOptic disc segmentationWeak labels

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

2.7K
Author Spotlight: Unraveling the Pathogenesis of Age-Related Macular Degeneration and Discovering Potential Therapies
06:16

Author Spotlight: Unraveling the Pathogenesis of Age-Related Macular Degeneration and Discovering Potential Therapies

Published on: July 28, 2023

2.7K

Related Experiment Videos

Last Updated: Sep 28, 2025

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

3.0K
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

2.7K
Author Spotlight: Unraveling the Pathogenesis of Age-Related Macular Degeneration and Discovering Potential Therapies
06:16

Author Spotlight: Unraveling the Pathogenesis of Age-Related Macular Degeneration and Discovering Potential Therapies

Published on: July 28, 2023

2.7K

Area of Science:

  • Ophthalmology
  • Medical Image Analysis
  • Machine Learning

Background:

  • Fundus image analysis is crucial for diagnosing ophthalmic diseases.
  • Optic disc segmentation is a key pre-processing step for analyzing pathological changes.
  • Current methods require time-consuming manual annotations, leading to inter-subject variance.

Purpose of the Study:

  • To develop a weakly supervised method for optic disc segmentation.
  • To overcome limitations of manual annotation in supervised learning for fundus image analysis.

Main Methods:

  • Proposed a weak label based Bayesian U-Net.
  • Utilized Hough transform for generating weak annotations.
  • Employed a probabilistic graphical model and Bayesian approach within the U-Net framework.
  • Optimized the model using the expectation-maximization algorithm.

Main Results:

  • Achieved strong performance in optic disc segmentation.
  • Demonstrated effectiveness compared to fully- and weakly-supervised baselines.
  • Reduced the need for pixel-level manual segmentation.

Conclusions:

  • The proposed Bayesian U-Net with weak labels offers an efficient and accurate solution for optic disc segmentation.
  • This method has the potential to streamline ophthalmic disease analysis by reducing annotation burden.
  • Weakly supervised learning is a viable approach for medical image segmentation tasks.