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

FAA-Net: Fetal abdominal anomaly diagnosis in prenatal ultrasound via LLM-enhanced multi-instance learning.

Medical image analysis·2026
Same author

Identification of preeclampsia-associated immune-related risk loci in the Chinese population.

Taiwanese journal of obstetrics & gynecology·2026
Same author

Artificial intelligence for detecting fetal orofacial clefts and advancing medical education.

Nature communications·2026
Same author

Multi-omic analysis of deep learning-derived phenotypes links ophthalmic imaging to cardiovascular and neurological traits.

Nature cardiovascular research·2026
Same author

From pixels to polygons: A survey of deep learning approaches for medical image-to-mesh reconstruction.

Medical image analysis·2026
Same author

A Deep Learning Model Integrating Ultrasound Images and Multidimensional Clinical Information for Differentiating Benign and Malignant Non-Mass Lesions.

Diagnostics (Basel, Switzerland)·2026
Same journal

SynTME: A tumor microenvironment-aware, pharmacology-inspired multi-stage framework for drug synergy prediction.

Computer methods and programs in biomedicine·2026
Same journal

MMFVS-Net: A triple-symmetric cross-attention network for multimodal optical image fusion and high-accuracy virtual staining of breast cancer tissues.

Computer methods and programs in biomedicine·2026
Same journal

A novel Milstein-stochastic epidemiologically-informed neural network for approaching epidemic dynamics: Application to Mpox disease.

Computer methods and programs in biomedicine·2026
Same journal

Accounting for approximation errors using surrogate-based parameter estimation of cardiac mechanics digital twins.

Computer methods and programs in biomedicine·2026
Same journal

Facial iPPG heatmap patterns based on period-aware autoencoder show association with carotid atherosclerosis towards non-contact hemodynamic assessment.

Computer methods and programs in biomedicine·2026
Same journal

Explainable machine learning models predict liver fibrosis risk and outcome in the general population: Development and multi-cohort external validation.

Computer methods and programs in biomedicine·2026
See all related articles

Related Experiment Video

Updated: Aug 5, 2025

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

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

470

Test-time bi-directional adaptation between image and model for robust segmentation.

Xiaoqiong Huang1, Xin Yang1, Haoran Dou2

  • 1National-Regional Key Technology Engineering Laboratory for Medical Ultrasound, School of Biomedical Engineering, Health Science Center, Shenzhen University, Shenzhen, China; Medical Ultrasound Image Computing (MUSIC) Laboratory, Shenzhen University, Shenzhen, China; Marshall Laboratory of Biomedidcal Engineering, Shenzhen University, Shenzhen, China.

Computer Methods and Programs in Biomedicine
|March 27, 2023
PubMed
Summary
This summary is machine-generated.

This study introduces a novel test-time adaptation method to improve deep learning model performance on medical images with unknown appearance shifts. The approach enhances segmentation accuracy in real-world clinical settings.

Keywords:
Medical image segmentationSelf-supervised learningStyle transferTest-time adaptation

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
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

2.6K

Related Experiment Videos

Last Updated: Aug 5, 2025

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

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

470
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
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

2.6K

Area of Science:

  • Medical image analysis
  • Deep learning in healthcare
  • Computer vision applications

Background:

  • Deep learning models face performance drops in clinical settings due to image appearance variations.
  • Existing adaptation methods often require target domain samples during training, which is impractical.
  • Unforeseen appearance shifts challenge the reliability of deployed medical imaging AI.

Purpose of the Study:

  • To develop a general method for robust segmentation of medical images with unknown appearance shifts.
  • To enhance the deployability of deep learning segmentation models in daily clinical practice.
  • To overcome limitations of training-time adaptation methods.

Main Methods:

  • Proposed a test-time bi-directional adaptation framework combining Image-to-Model (I2M) and Model-to-Image (M2I) strategies.
  • I2M adapts test images to the model using statistical alignment style transfer.
  • M2I adapts the model to test images via self-supervised learning with proxy labels and a proxy consistency criterion.

Main Results:

  • Demonstrated robust segmentation performance against unknown appearance shifts across 10 diverse datasets.
  • Achieved promising robustness and efficiency in segmenting fetal ultrasound, chest X-ray, and retinal fundus images.
  • Validated the effectiveness of the complementary I2M and M2I framework.

Conclusions:

  • Addressed the critical challenge of appearance shifts in clinical medical images.
  • Provided a general and robust solution for medical image segmentation applicable in clinical settings.
  • The proposed method enhances the reliability of AI tools in real-world healthcare scenarios.