Leveraging redundancy in simultaneous multislice acquisitions to improve spike detection

Area of Science:

• Medical Imaging
• Magnetic Resonance Imaging (MRI)
• Image Quality Assurance

Background:

• Spike noise artifacts can compromise MRI data quality.
• Daily quality assurance (QA) protocols are essential for reliable MRI scans.
• Simultaneous multislice (SMS) acquisitions offer potential for improved artifact detection due to data redundancy.

Purpose of the Study:

• To enhance the detection of low-level spike noise artifacts in MRI.
• To leverage the redundancy in simultaneous multislice (SMS) acquisitions for improved artifact detection.
• To develop a more efficient daily QA protocol for MRI.

Main Methods:

• Magnitude MRI images were converted into pseudo k-space representations.
• Time series data at each pseudo k-space point were detrended.
• A slice was flagged for artifact if it exceeded an intensity threshold and simultaneously acquired slices showed outliers (SMS criterion).

Main Results:

• Out of 401,112 inspected slices, 42 contained spike artifacts.
• The SMS criterion correctly identified all artifact-containing slices with only 30 false positives.
• Without the SMS criterion, 12,908 slices were incorrectly flagged, necessitating a trade-off between sensitivity and specificity.

Conclusions:

• The SMS criterion effectively reduces false positives in spike artifact detection.
• Accurate detection of low-level spike artifacts enables early identification of hardware issues.
• This method streamlines daily QA, avoiding time-intensive retrospective data correction.