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: Nov 6, 2025

Deep Neural Networks for Image-Based Dietary Assessment

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

Scale-adaptive deep network for deformable image registration.

Yudi Sang¹, Dan Ruan¹

¹Department of Bioengineering and Department of Radiation Oncology, University of California, Los Angeles, CA, USA.

Medical Physics

|May 12, 2021

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

Beyond the LUMIR challenge: The pathway to foundational registration models.

Medical image analysis·2026

Same author

Free-Breathing Dynamic, Regularized, Adaptive Cluster Optimization (DRACO) Cine Cardiac MRI in Atrial Fibrillation.

Journal of magnetic resonance imaging : JMRI·2026

Same author

Survey of work-from-home experiences among medical physicists in Southern California during and after the COVID-19 pandemic.

Journal of applied clinical medical physics·2026

Same author

Hierarchical heated markov modeling for synthesizing activity data from wearable devices.

Computational and structural biotechnology journal·2026

Same author

Benchmark of Segmentation Techniques for Pelvic Fracture in CT and X-Ray: Summary of the PENGWIN 2024 Challenge.

IEEE transactions on medical imaging·2026

Same author

CIP2A mediates mitotic recruitment of SLX4/MUS81/XPF to resolve replication stress-induced DNA lesions.

Nature communications·2025

Same journal

Decomposition-based harmonization for quantitative PET imaging across scanners and radiotracers.

Medical physics·2026

Same journal

Development and evaluation of an in vivo dose-based monitoring system for electron FLASH radiation therapy.

Medical physics·2026

Same journal

A novel optical respiratory gating system with a hybrid phase-amplitude algorithm for spot-scanning proton therapy.

Medical physics·2026

Same journal

Gamma Knife treatment planning using knowledge-based reinforcement learning.

Medical physics·2026

Same journal

Development and characterization of a novel, small animal external beam irradiator using a clinical high dose rate brachytherapy source.

Medical physics·2026

Same journal

Deep learning-based dose prediction for MR-guided prostate SIB: Supporting rapid feasibility assessment and adaptive editing margin selection.

Medical physics·2026

See all related articles

This study introduces a novel image registration network using Dilated Inception Modules (DIMs) and Scale Adaptive Modules (SAMs) for efficient, multi-scale deformation handling. The proposed method significantly improves registration accuracy and speed compared to existing techniques.

Area of Science:

Medical Imaging
Computer Vision
Machine Learning

Background:

Conventional image registration uses multiresolution strategies to handle deformations and avoid local minima.
Deep networks capture scale via receptive fields, but simultaneous multi-scale feature extraction remains challenging.

Purpose of the Study:

To propose a novel image registration network that is conscious of and self-adaptive to deformations of various scales.
To improve overall image registration performance by addressing the heterogeneous scale problem.

Main Methods:

Proposed Dilated Inception Modules (DIMs) to efficiently incorporate receptive fields of different sizes.
Introduced Scale Adaptive Modules (SAMs) to guide shallow features using spatially adaptive dilation rates learned from deep features.

Keywords:

deep learning dilated convolution image registration scale self-adaptive

More Related Videos

Multi-modal Pulmonary Imaging: Using Complementary Information from CT and Hyperpolarized 129Xe MRI to Evaluate Lung Structure-Function

Multi-modal Pulmonary Imaging: Using Complementary Information from CT and Hyperpolarized 129Xe MRI to Evaluate Lung Structure-Function

Published on: April 12, 2024

Related Experiment Videos

Last Updated: Nov 6, 2025

Deep Neural Networks for Image-Based Dietary Assessment

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

Multi-modal Pulmonary Imaging: Using Complementary Information from CT and Hyperpolarized 129Xe MRI to Evaluate Lung Structure-Function

Multi-modal Pulmonary Imaging: Using Complementary Information from CT and Hyperpolarized 129Xe MRI to Evaluate Lung Structure-Function

Published on: April 12, 2024

Integrated DIMs and SAMs into a U-net based registration network trained in an unsupervised setting for single-evaluation registration.

Main Results:

Experiments on 2D cardiac MRIs showed SAM's adaptive dilation rate correlated well with deformation scale, achieving a Dice coefficient of (0.93 ± 0.02) for left ventricle segmentation, outperforming existing methods.
Synthetic data experiments demonstrated significant reduction in Target Registration Error (TRE) using DIMs and SAMs.
The 3D network achieved a mean TRE of 2.52 mm on thoracic 4DCTs, significantly lower than baseline methods, with rapid registration times (0.002s for 2D, 0.42s for 3D).

Conclusions:

DIMs and SAMs efficiently and adaptively address the heterogeneous scale problem in image registration.
The proposed method offers an efficient alternative to traditional multiresolution registration setups.