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

Anatomy-Guided Radiology Report Generation With Pathology-Aware Regional Prompts.

IEEE open journal of engineering in medicine and biology·2026
Same author

Multimodal AI-driven analysis in breast and lung cancer: insights from the OPTIMA prototyping workshop.

BMC proceedings·2026
Same author

Non-invasive assessment of microcirculatory resistance by coronary computed tomography angiography.

EuroIntervention : journal of EuroPCR in collaboration with the Working Group on Interventional Cardiology of the European Society of Cardiology·2026
Same author

Reason like a radiologist: Chain-of-thought and reinforcement learning for verifiable report generation.

Medical image analysis·2025
Same author

Dynamical multi-order responses and global semantic-infused adversarial learning: A robust airway segmentation method.

Medical image analysis·2025
Same author

Confounder-aware foundation modeling for accurate phenotype profiling in cell imaging.

Npj imaging·2025

Related Experiment Video

Updated: Aug 3, 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

2.8K

Less Is More: Unsupervised Mask-Guided Annotated CT Image Synthesis With Minimum Manual Segmentations.

Xiaodan Xing, Giorgos Papanastasiou, Simon Walsh

    IEEE Transactions on Medical Imaging
    |April 8, 2023
    PubMed
    Summary

    This study introduces Unsupervised Mask (UM)-guided synthesis for medical image analysis, generating realistic synthetic images and segmentations with limited labels. This method improves data augmentation efficacy by enhancing fidelity and variety compared to previous approaches.

    More Related Videos

    Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
    14:08

    Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

    Published on: April 13, 2013

    42.7K
    Lung CT Segmentation to Identify Consolidations and Ground Glass Areas for Quantitative Assesment of SARS-CoV Pneumonia
    08:05

    Lung CT Segmentation to Identify Consolidations and Ground Glass Areas for Quantitative Assesment of SARS-CoV Pneumonia

    Published on: December 19, 2020

    14.2K

    Related Experiment Videos

    Last Updated: Aug 3, 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

    2.8K
    Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images
    14:08

    Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

    Published on: April 13, 2013

    42.7K
    Lung CT Segmentation to Identify Consolidations and Ground Glass Areas for Quantitative Assesment of SARS-CoV Pneumonia
    08:05

    Lung CT Segmentation to Identify Consolidations and Ground Glass Areas for Quantitative Assesment of SARS-CoV Pneumonia

    Published on: December 19, 2020

    14.2K

    Area of Science:

    • Medical Image Analysis
    • Deep Learning
    • Artificial Intelligence

    Background:

    • Data augmentation via synthesis improves deep learning performance in medical imaging.
    • Generating paired segmentation masks for synthetic images is challenging and subjective.
    • Existing conditional generative models require extensive labeled data and offer limited anatomical control, leading to unrealistic features and reduced augmentation utility.

    Purpose of the Study:

    • To develop a novel Unsupervised Mask (UM)-guided synthesis strategy for generating paired synthetic medical images and segmentations.
    • To overcome limitations of existing methods, including reliance on large labeled datasets and poor anatomical realism.
    • To enhance the fidelity, variety, and utility of synthetic medical data for improved deep learning applications.

    Main Methods:

    • Developed a superpixel-based algorithm for unsupervised structural guidance generation.
    • Designed a conditional generative model for simultaneous synthesis of images and annotations in a semi-supervised multi-task setting.
    • Introduced multi-scale multi-task Fréchet Inception Distance (MM-FID) and multi-scale multi-task standard deviation (MM-STD) for robust evaluation of synthetic image quality.

    Main Results:

    • UM-guided synthesis successfully generated high-quality synthetic images and segmentations using limited manual labels.
    • The proposed method demonstrated significantly higher fidelity, variety, and utility compared to segmentation mask-guided synthesis.
    • MM-FID and MM-STD provided stable and reproducible image quality measurements across different scales.

    Conclusions:

    • Unsupervised Mask (UM)-guided synthesis offers a powerful and efficient approach for medical image data augmentation.
    • The method addresses key limitations of previous synthesis techniques, enabling more realistic and diverse synthetic data generation.
    • This advancement holds significant potential for improving deep learning model performance in medical image analysis.