Related Experiment Video
Updated: Sep 22, 2025
07:21
Magnetic Resonance Derived Myocardial Strain Assessment Using Feature Tracking
Published on: February 12, 2011
14.5K
Motion correction for native myocardial T1 mapping using self-supervised deep learning registration with contrast
Yuze Li1, Chunyan Wu1, Haikun Qi2
1Center for Biomedical Imaging Research (CBIR), School of Medicine, Tsinghua University, Beijing, China.
NMR in Biomedicine
|May 23, 2022
Summary
This study introduces a novel self-supervised deep learning method (SDRAP) to correct motion in cardiac T1 mapping. SDRAP significantly improves accuracy and image quality, offering a faster alternative for clinical applications.
Area of Science:
- Medical Imaging
- Cardiovascular MRI
- Artificial Intelligence in Medicine
Background:
- Motion artifacts are a major challenge in myocardial T1 mapping, potentially leading to inaccurate T1 estimation and misdiagnosis.
- Accurate T1 mapping is crucial for diagnosing and monitoring various cardiac conditions.
Purpose of the Study:
- To develop and evaluate a novel motion correction method for myocardial T1 mapping using self-supervised deep learning with contrast separation (SDRAP).
- To compare the performance of SDRAP against traditional registration methods in terms of accuracy, image quality, and computational efficiency.
Main Methods:
- Proposed a sparse coding method to separate contrast from T1-weighted (T1w) images.
- Developed a self-supervised deep neural network (SDRAP-CC) utilizing cross-correlation for image registration.
- Applied signal fitting to motion-corrected T1w images to generate accurate T1 maps using MOLLI sequence data from 80 healthy volunteers.
Main Results:
- SDRAP-CC achieved superior myocardium contour delineation with a Dice similarity coefficient (DSC) of 85.0 ± 3.9% and a mean boundary error (MBE) of 0.92 ± 0.25 mm.
- SDRAP-CC resulted in lower T1 value standard deviation (28.1 ± 17.6 ms) and improved subjective image quality scores.
- The SDRAP method demonstrated significant acceleration, registering images in 0.52 seconds per slice compared to 3.7 seconds for FFD.
Conclusions:
- The proposed SDRAP method, particularly SDRAP-CC, effectively corrects motion artifacts in myocardial T1 mapping.
- SDRAP offers improved accuracy, enhanced image quality, and substantial computational speedup, making it a promising tool for clinical cardiovascular MRI.
