Search research articles

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: Dec 29, 2025

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

Unpaired Multi-Modal Segmentation via Knowledge Distillation.

Qi Dou, Quande Liu, Pheng Ann Heng

IEEE Transactions on Medical Imaging

|February 4, 2020

Summary

This summary is machine-generated.

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

Disparate privacy risks from medical AI.

Nature·2026

Same author

Optical metasurfaces for general vision processing on the edge.

Nature·2026

Same author

Short-Term, Cycle-Synchronized Adjunctive Therapy With Crocus Total Glucosides Tablets for Cardiac Protection Against Cancer Therapy-Related Cardiac Dysfunction in Breast Cancer Patients: A Randomized, Double-Blind, Placebo-Controlled Trial.

MedComm·2026

Same author

DSTED: decoupling temporal stabilization and discriminative enhancement for surgical workflow recognition.

International journal of computer assisted radiology and surgery·2026

Same author

Learning the PTM code through a coarse-to-fine mechanism-aware framework.

Nature communications·2026

Same author

Current validation practice undermines surgical AI development.

ArXiv·2026

Same journal

Physiology-guided Self-supervised Learning for Simultaneous Dual-Tracer PET Separation.

IEEE transactions on medical imaging·2026

Same journal

Informed-Exploration Reinforcement Learning for Automated Virtual Coronary Intervention Planning.

IEEE transactions on medical imaging·2026

Same journal

4D Reconstruction of Fetal Left Ventricle from Echocardiography via 2.5D Radial Segmentation and Graph-Fourier Reconstruction.

IEEE transactions on medical imaging·2026

Same journal

Generalised Medical Phrase Grounding.

IEEE transactions on medical imaging·2026

Same journal

EndoLRMGS: Combining Large Reconstruction Modelling and Gaussian Splatting for Complete Endoscopic Scene Reconstruction.

IEEE transactions on medical imaging·2026

Same journal

A Neural-Analytical Fusion Scatter Correction Method for Multi-Source CT Using Equivalent High-Order Scatter.

IEEE transactions on medical imaging·2026

See all related articles

This study introduces a novel, compact multi-modal learning scheme for medical image segmentation, significantly improving accuracy by sharing parameters and using a knowledge distillation-inspired loss. The method effectively segments cardiac and abdominal structures across CT and MRI modalities.

Area of Science:

Medical imaging
Artificial intelligence
Computer vision

Background:

Multi-modal learning in medical imaging often requires complex architectures with modality-specific layers.
Co-registered images from different modalities (e.g., CT and MRI) are typically used for training.
Existing methods may not fully leverage parameter sharing for compact and efficient models.

Purpose of the Study:

To propose a novel, compact learning scheme for unpaired cross-modality image segmentation.
To achieve superior segmentation accuracy using a highly compact network architecture.
To enable effective training of shared parameter models through a new loss function.

Main Methods:

A novel learning scheme with a highly compact architecture is proposed, heavily reusing network parameters by sharing all convolutional kernels across CT and MRI.

More Related Videos

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

Related Experiment Videos

Last Updated: Dec 29, 2025

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

Modality-specific internal normalization layers are employed to compute respective statistics.

A novel loss term inspired by knowledge distillation, constraining KL-divergence of prediction distributions between modalities, is introduced for effective training.

Main Results:

The proposed method consistently outperforms single-modal training and previous multi-modal approaches on cardiac and abdominal organ segmentation tasks.
Validation was performed on two multi-class segmentation problems using 2D dilated networks and 3D U-Net architectures.
The compact architecture with shared kernels and modality-specific normalization achieved superior segmentation accuracy.

Conclusions:

The novel multi-modal learning scheme offers a highly effective and compact approach for cross-modality medical image segmentation.
Parameter sharing and a knowledge distillation-inspired loss significantly enhance segmentation performance.
The method demonstrates general efficacy across different network settings and segmentation tasks.