A survey on deep learning in medical image registration: new technologies, uncertainty, evaluation metrics, and beyond

Junyu Chen,Yihao Liu,Shuwen Wei,Zhangxing Bian,Shalini Subramanian,Aaron Carass,Jerry L. Prince,Yong Du

2024-11-01

Abstract:Deep learning technologies have dramatically reshaped the field of medical image registration over the past decade. The initial developments, such as regression-based and U-Net-based networks, established the foundation for deep learning in image registration. Subsequent progress has been made in various aspects of deep learning-based registration, including similarity measures, deformation regularizations, network architectures, and uncertainty estimation. These advancements have not only enriched the field of image registration but have also facilitated its application in a wide range of tasks, including atlas construction, multi-atlas segmentation, motion estimation, and 2D-3D registration. In this paper, we present a comprehensive overview of the most recent advancements in deep learning-based image registration. We begin with a concise introduction to the core concepts of deep learning-based image registration. Then, we delve into innovative network architectures, loss functions specific to registration, and methods for estimating registration uncertainty. Additionally, this paper explores appropriate evaluation metrics for assessing the performance of deep learning models in registration tasks. Finally, we highlight the practical applications of these novel techniques in medical imaging and discuss the future prospects of deep learning-based image registration.

Image and Video Processing,Computer Vision and Pattern Recognition

What problem does this paper attempt to address?

This paper aims to solve several key problems in the field of medical image registration:

1. **Technological Advances**: The paper reviews the development of deep - learning - based medical image registration techniques in the past decade, especially how early methods such as regression networks and U - Net have laid the foundation for the application of deep learning in image registration. Subsequently, the paper discusses the progress made in similarity measurement, deformation regularization, network architecture, and uncertainty estimation.

2. **Performance Evaluation**: The paper explores evaluation metrics suitable for assessing the performance of deep - learning models in registration tasks, which helps researchers and clinicians better understand and compare the effects of different methods.

3. **Practical Applications**: The paper emphasizes the practical applications of these new technologies in medical imaging, including tasks such as constructing anatomical atlases, multi - atlas segmentation, motion estimation, and 2D - 3D registration.

4. **Future Prospects**: The paper also discusses the future research directions and challenges of deep - learning - based medical image registration, especially further research in uncertainty estimation and evaluation metrics.

Through the exploration of these issues, the paper provides readers with a comprehensive perspective to understand the latest progress in the current field of deep - learning - based medical image registration and its potential application prospects.