Search research articles

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

Brain network pathophysiology in dystonia.

Dystonia (Lausanne, Switzerland)·2026

Same author

A scoping review of traditional and artificial intelligence methods in malaria diagnostics.

NPJ digital medicine·2026

Same author

CROSS-MODAL FINE-TUNING OF 3D CONVOLUTIONAL FOUNDATION MODELS FOR ADHD CLASSIFICATION WITH LOW-RANK ADAPTATION.

Proceedings. IEEE International Symposium on Biomedical Imaging·2026

Same author

DSHARP: Deep Incompressible Motion Estimation with Sinusoidal-transformed Harmonic Phase for Tagged MRI.

IEEE transactions on medical imaging·2026

Same author

Complexity of resting cortical activity predicts neurophysiological responses to theta-burst stimulation but fails to generalize: A rigorous machine-learning approach.

PLoS computational biology·2026

Same author

A speech-to-video synthesis approach using spatio-temporal diffusion for vocal tract MRI.

Medical image analysis·2026

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: Jul 24, 2025

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping

Published on: September 25, 2019

Incremental Learning for Heterogeneous Structure Segmentation in Brain Tumor MRI.

Xiaofeng Liu¹, Helen A Shih², Fangxu Xing¹

¹Gordon Center for Medical Imaging, Department of Radiology, Massachusetts General Hospital and Harvard Medical School, Boston, MA, 02114.

|July 3, 2023

Summary

This summary is machine-generated.

This study introduces a novel deep learning approach for continuous medical image segmentation, enabling models to adapt to new data and structures without forgetting previous knowledge. This lifelong learning framework ensures robust performance in evolving environments.

More Related Videos

Automated Segmentation of Cortical Grey Matter from T1-Weighted MRI Images

Automated Segmentation of Cortical Grey Matter from T1-Weighted MRI Images

Published on: January 7, 2019

Author Spotlight: Bridging Gaps in Anatomy and Establishing a Foundation for Algorithmic Studies

Author Spotlight: Bridging Gaps in Anatomy and Establishing a Foundation for Algorithmic Studies

Published on: December 15, 2023

Related Experiment Videos

Last Updated: Jul 24, 2025

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping

Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping

Published on: September 25, 2019

Automated Segmentation of Cortical Grey Matter from T1-Weighted MRI Images

Automated Segmentation of Cortical Grey Matter from T1-Weighted MRI Images

Published on: January 7, 2019

Author Spotlight: Bridging Gaps in Anatomy and Establishing a Foundation for Algorithmic Studies

Author Spotlight: Bridging Gaps in Anatomy and Establishing a Foundation for Algorithmic Studies

Published on: December 15, 2023

Area of Science:

Medical image analysis
Artificial intelligence in healthcare
Computer vision

Background:

Static deep learning models struggle with evolving data in medical image segmentation.
Incremental learning is crucial for adapting models to new datasets and structures without performance degradation.
Distribution shifts and novel structures pose significant challenges for existing segmentation models.

Approach:

Developed a divergence-aware dual-flow module with balanced rigidity and plasticity for task decoupling.
Implemented a pseudo-label training scheme with self-entropy regularized momentum MixUp decay for adaptive optimization.
Proposed a unified framework to progressively evolve off-the-shelf segmentation models to diverse datasets with new anatomical categories.

Key Points:

The framework effectively decouples old and new tasks using a dual-flow module guided by continuous batch renormalization.
Pseudo-labeling and adaptive optimization strategies enhance network performance on continually changing target domains.
The model successfully retained discriminability of previously learned structures during brain tumor segmentation tasks.

Conclusions:

The proposed method enables realistic lifelong extension of segmentation models with accumulating big medical data.
This approach addresses catastrophic forgetting and distribution shifts in incremental learning settings.
Achieved robust and adaptable deep learning-based segmentation for evolving medical imaging environments.