Related Experiment Video
Updated: Dec 25, 2025
04:17
DNA Virus Detection System Based on RPA-CRISPR/Cas12a-SPM and Deep Learning
Published on: May 10, 2024
1.3K
Overview of quantitative susceptibility mapping using deep learning: Current status, challenges and opportunities.
Woojin Jung1, Steffen Bollmann2,3, Jongho Lee1
1Laboratory for Imaging Science and Technology, Department of Electrical and Computer Engineering, Seoul National University, Seoul, South Korea.
NMR in Biomedicine
|March 25, 2020
Summary
Quantitative susceptibility mapping (QSM) extracts tissue magnetic susceptibility using MRI. Deep learning methods offer efficient QSM processing, bypassing complex iterative optimization and regularization parameter selection for disease diagnosis.
Area of Science:
- Biomedical Imaging
- Neuroimaging
- Medical Physics
Background:
- Quantitative susceptibility mapping (QSM) measures tissue magnetic susceptibility from MRI phase data.
- QSM is influenced by myelin, iron, and calcium, offering insights into various diseases.
- Current QSM methods involve computationally intensive iterative optimization for processing steps.
Purpose of the Study:
- To review the application of deep learning (DL) in quantitative susceptibility mapping (QSM).
- To highlight the advantages and limitations of DL-based QSM processing.
- To discuss future research directions for DL in QSM.
Main Methods:
- Review of existing literature on deep learning approaches for QSM.
- Focus on convolutional neural networks (CNNs) for QSM reconstruction.
- Comparison of DL-based methods with traditional iterative optimization techniques.
Main Results:
- Deep learning models can solve QSM processing steps (phase unwrapping, background field removal, inversion) via feed-forward computations.
- DL approaches eliminate the need for iterative optimization and regularization parameter tuning.
- Demonstrated efficiency and potential of DL for QSM reconstruction.
Conclusions:
- Deep learning significantly enhances the efficiency of quantitative susceptibility mapping.
- DL-based QSM offers a promising alternative to conventional, computationally demanding methods.
- Further research is needed to fully exploit DL advancements for QSM applications in disease.
