Related Experiment Video
Updated: Jul 6, 2025
08:51
Magnetic Resonance Imaging of Multiple Sclerosis at 7.0 Tesla
Published on: February 19, 2021
9.0K
Maximum spherical mean value filtering for whole-brain QSM
Alexandra G Roberts1,2, Dominick J Romano2,3, Mert Şişman1,2
1Department of Electrical and Computer Engineering, Cornell University, Ithaca, New York, USA.
Magnetic Resonance in Medicine
|January 3, 2024
Summary
A new maximum spherical mean value (mSMV) algorithm reduces brain QSM shadow artifacts without tissue erosion. This method improves QSM accuracy and image quality in simulations and human studies.
Area of Science:
- Medical imaging
- Biophysics
- Neuroscience
Background:
- Background field is a primary cause of shadow artifacts in Quantitative Susceptibility Mapping (QSM).
- Existing QSM algorithms may require brain-tissue erosion to mitigate these artifacts.
Purpose of the Study:
- To introduce the maximum spherical mean value (mSMV) algorithm for QSM.
- To reduce shadow artifacts in brain QSM without eroding tissue volume.
Main Methods:
- Developed the mSMV algorithm to filter large field-magnitude values near the border.
- Evaluated mSMV effectiveness using numerical brain simulations.
- Assessed mSMV performance with in vivo human data from healthy volunteers and patients.
Main Results:
- mSMV effectively reduced shadow artifacts and enhanced QSM accuracy in simulations.
- Observed improved shadow reduction in healthy subjects and patients with various neurological conditions.
- Demonstrated lower QSM variation in gray matter and higher image quality scores with mSMV.
Conclusions:
- The mSMV algorithm provides QSM maps comparable to SMV-filtered dipole inversion.
- mSMV successfully reduces artifacts without compromising the volume of interest.
