Book chapter
Get the image: Machine learning for MR image reconstruction
Machine Learning in MRI: From Methods to Clinical Translation, pp.189-218
Advances in Magnetic Resonance Technology and Applications, v. 13, Academic Press
2025
DOI: 10.1016/B978-0-443-14109-6.00007-9
Abstract
This chapter provides an in-depth exploration of artificial intelligence (AI) techniques for solving inverse problems of image reconstruction in magnetic resonance imaging (MRI). It begins with the fundamentals of inverse problems in MRI, discussing both linear and non-linear models, and presents classical reconstruction methods alongside advanced machine learning techniques. Topics include deep learning architectures, self-supervised learning frameworks, and generative models such as GANs and diffusion models. Practical applications, such as motion correction, low-field MRI enhancement, and quantitative imaging, are reviewed. Finally, critical challenges such as model robustness, bias, and generalization issues are discussed, demonstrating current limitations of AI methods and areas in needs for further research.
Details
- Title: Subtitle
- Get the image: Machine learning for MR image reconstruction
- Creators
- Shanshan Wang - Shenzhen Institutes of Advanced TechnologyRuoyou Wu - Shenzhen Institutes of Advanced TechnologyReinhard Heckel - Technical University of MunichMathews Jacob - Department of Electrical and Computer Engineering, University of Lowa, Lowa, IA, United StatesEfrat Shimron - Technion – Israel Institute of Technology
- Resource Type
- Book chapter
- Publication Details
- Machine Learning in MRI: From Methods to Clinical Translation, pp.189-218
- Series
- Advances in Magnetic Resonance Technology and Applications; v. 13
- DOI
- 10.1016/B978-0-443-14109-6.00007-9
- ISSN
- 2666-9099
- Publisher
- Academic Press; Amsterdam
- Language
- English
- Date published
- 2025
- Academic Unit
- Electrical and Computer Engineering
- Record Identifier
- 9985097023902771
Metrics
1 Record Views