Related Experiment Video
Updated: Dec 25, 2025
04:48
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
Published on: November 30, 2022
3.3K
Deep learning in medical image registration: a review.
Yabo Fu1, Yang Lei1, Tonghe Wang1,2
1Department of Radiation Oncology, Emory University, Atlanta, GA, United States of America.
Physics in Medicine and Biology
|March 29, 2020
Summary
This review explores deep learning (DL) for medical image registration, categorizing methods and assessing their potential. It highlights key challenges and future trends in DL-based medical imaging.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Computer Vision
Background:
- Medical image registration is crucial for diagnosis and treatment planning.
- Traditional methods face limitations in accuracy and efficiency.
- Deep learning (DL) offers promising advancements in medical image registration.
Purpose of the Study:
- To review and categorize DL-based medical image registration methods.
- To highlight recent developments, applications, and challenges in the field.
- To analyze trends and future potential of DL in medical image registration.
Main Methods:
- Systematic review of DL-based medical image registration literature.
- Classification of methods into seven categories based on techniques, functions, and popularity.
- Comparative analysis of DL methods for lung and brain registration using benchmark datasets.
Main Results:
- Categorization of DL registration methods into seven distinct groups.
- Identification of key contributions, challenges, and future potential for each category.
- Comprehensive comparison of DL methods for lung and brain imaging, with statistical analysis of cited works.
Conclusions:
- Deep learning is rapidly advancing medical image registration.
- Understanding current methods and challenges is vital for future research.
- DL-based registration shows significant potential for improving clinical workflows.
