Semiautomated method for noise reduction and background phase error correction in MR phase velocity data

Area of Science:

• Medical Imaging
• Cardiovascular Physiology
• Biomedical Engineering

Background:

• Magnetic resonance (MR) phase velocity measurements are crucial for assessing blood flow.
• Background phase distortion and random noise degrade the quality and accuracy of these measurements.
• Accurate flow quantification is essential for diagnosing cardiovascular conditions.

Purpose of the Study:

• To develop and demonstrate a semiautomated method for reducing background phase distortion and random noise in MR phase velocity measurements.
• To improve the quality of phase velocity images for better cardiovascular assessment.
• To validate the technique using human left ventricular outflow tract data.

Main Methods:

• Calculated time standard deviations of phase velocity images over a cardiac cycle.
• Identified static regions with low standard deviation to approximate background distortion.
• Fitted a flat surface to static regions for distortion correction and applied it to phase velocity images.
• Removed random noise by zeroing regions with high standard deviation.

Main Results:

• Successfully reduced background phase distortion and random noise in MR phase velocity images.
• Demonstrated the technique's effectiveness on human left ventricular outflow tract data.
• Presented corrected velocity data in vector and contour forms superimposed on MR angiographic images.

Conclusions:

• The developed semiautomated method effectively corrects phase distortion and reduces noise in MR phase velocity measurements.
• This technique enhances the reliability of quantitative flow imaging in cardiovascular applications.
• The method offers a valuable tool for improving the diagnostic accuracy of MR-based cardiac flow assessment.