A visual quality control scale for clinical arterial spin labeling images

Area of Science:

• Radiology
• Medical Imaging
• Neuroimaging

Background:

• Image quality assessment is crucial for accurate clinical evaluation of MRI scans.
• Arterial Spin Labeling (ASL) provides vital information on cerebral blood flow (CBF).
• Standardized quality control is needed for ASL and CBF map interpretation.

Purpose of the Study:

• To introduce a visual scoring system for ASL image quality control.
• To establish a standard for evaluating ASL and cerebral blood flow (CBF) maps.
• To assess the reproducibility and reliability of the proposed visual scoring system.

Main Methods:

• A two-component visual quality control (QC) system was developed: contrast-based QC (cQC) and artifact-based QC (aQC).
• Three raters evaluated 158 QUASAR ASL scans using cQC and aQC.
• Statistical analyses included Spearman correlation and intraclass correlation coefficients (ICC).

Main Results:

• High intra-rater (0.80-0.90) and inter-rater (0.60-0.74) agreement was observed for the QC system.
• Strong correlations were found between artifact-based QC and CBF map quality (r=0.75).
• A QC score of 18 was identified as optimal for differentiating high-quality scans.

Conclusions:

• The proposed visual QC system demonstrates high reproducibility for ASL images.
• The system provides a reliable threshold for discarding low-quality ASL scans.
• Future work should compare this visual QC system with automated methods.