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

Copper-mediated amidation of alkenylzirconocenes with acyl azides: formation of enamides.

Organic letters·2013
Same author

JARID1A, JMY, and PTGER4 polymorphisms are related to ankylosing spondylitis in Chinese Han patients: a case-control study.

PloS one·2013
Same author

[The risk factors of ventilator-associated pneumonia in newborn and the changes of isolated pathogens].

Sichuan da xue xue bao. Yi xue ban = Journal of Sichuan University. Medical science edition·2013
Same author

A route to phase controllable Cu2ZnSn(S(1-x)Se(x))4 nanocrystals with tunable energy bands.

Scientific reports·2013
Same author

Efficacy of an infection control program in reducing ventilator-associated pneumonia in a Chinese neonatal intensive care unit.

American journal of infection control·2013
Same author

[Effect of different forms of inorganic nitrogen on the photodegradation of antipyrine in water].

Huan jing ke xue= Huanjing kexue·2013

Related Experiment Video

Updated: Jun 30, 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

2.7K

Anomaly-guided weakly supervised lesion segmentation on retinal OCT images.

Jiaqi Yang1, Nitish Mehta2, Gozde Demirci1

  • 1Graduate Center CUNY, 365 5th Ave, NY 10016, USA.

Medical Image Analysis
|March 17, 2024
PubMed
Summary

This study introduces a novel anomaly-guided mechanism (AGM) for multi-class segmentation in retinal optical coherence tomography (OCT) images using only image-level labels, improving lesion detection accuracy.

Keywords:
Anomaly detectionMulti-label classificationRetinal OCT lesion segmentationSelf-attentionWeakly supervised segmentation

More Related Videos

Multimodal Volumetric Retinal Imaging by Oblique Scanning Laser Ophthalmoscopy oSLO and Optical Coherence Tomography OCT
12:22

Multimodal Volumetric Retinal Imaging by Oblique Scanning Laser Ophthalmoscopy oSLO and Optical Coherence Tomography OCT

Published on: August 4, 2018

8.5K
Application of Optical Coherence Tomography to a Mouse Model of Retinopathy
08:22

Application of Optical Coherence Tomography to a Mouse Model of Retinopathy

Published on: January 12, 2022

4.0K

Related Experiment Videos

Last Updated: Jun 30, 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

2.7K
Multimodal Volumetric Retinal Imaging by Oblique Scanning Laser Ophthalmoscopy oSLO and Optical Coherence Tomography OCT
12:22

Multimodal Volumetric Retinal Imaging by Oblique Scanning Laser Ophthalmoscopy oSLO and Optical Coherence Tomography OCT

Published on: August 4, 2018

8.5K
Application of Optical Coherence Tomography to a Mouse Model of Retinopathy
08:22

Application of Optical Coherence Tomography to a Mouse Model of Retinopathy

Published on: January 12, 2022

4.0K

Area of Science:

  • Biomedical research
  • Medical imaging
  • Computer vision

Background:

  • Big data in biomedical research requires expert labeling for supervised learning.
  • Pixel-level annotation is labor-intensive and costly.
  • Existing weakly supervised semantic segmentation (WSSS) methods often focus on single-class segmentation and struggle with multi-class medical images.

Purpose of the Study:

  • To develop a novel weakly supervised semantic segmentation (WSSS) method for multi-class segmentation in retinal optical coherence tomography (OCT) images.
  • To address the limitations of existing WSSS methods in handling multiple co-existing classes and variations in lesion scale and occurrence in medical imaging.
  • To achieve accurate lesion segmentation using only image-level labels.

Main Methods:

  • Proposed a novel anomaly-guided mechanism (AGM) for multi-class segmentation.
  • Leveraged anomaly detection and self-attention to integrate weak abnormal signals with global contextual information.
  • Incorporated an iterative refinement stage to enhance focus on potential lesions and suppress irrelevant regions.

Main Results:

  • The proposed AGM achieved state-of-the-art performance in WSSS for lesion segmentation on OCT images.
  • Validated performance across two public and one private dataset, demonstrating robustness.
  • Effectively handled multi-class segmentation challenges in medical images with significant variations.

Conclusions:

  • The anomaly-guided mechanism (AGM) offers a promising solution for accurate multi-class lesion segmentation in OCT images using only image-level labels.
  • This approach overcomes limitations of existing WSSS methods in medical imaging, particularly for complex, multi-class scenarios.
  • The method holds potential for advancing automated analysis of retinal diseases from OCT scans.