Reproducibility of quantitative structural and physiological MRI measurements

Area of Science:

• Neuroimaging
• Quantitative MRI/S

Background:

• Longitudinal quantitative magnetic resonance imaging and spectroscopy (MRI/S) is crucial for assessing neurological disorders and treatment efficacy.
• Understanding the inherent technical and physiological consistency of MRI/S measurements is vital for accurate interpretation.

Purpose of the Study:

• To examine the variance and reproducibility of commonly used quantitative MRI/S measurements in healthy subjects.
• To control for physiological and technical parameters during longitudinal imaging.
• To establish a baseline for interpreting MRI/S data in clinical contexts.

Main Methods:

• Twenty-five healthy subjects underwent three imaging sessions over five days on a 3T Siemens Verio scanner.
• Structural (T1, T2-weighted, diffusion-weighted imaging) and physiological (arterial spin labeling, proton magnetic resonance spectroscopy) data were acquired.
• Reproducibility was assessed using mean relative difference, coefficient of variation, and intraclass correlation (ICC), alongside a "reproducibility rating".

Main Results:

• Most structural MRI/S measurements demonstrated excellent reproducibility (ICCs 0.872-0.998).
• Moderate-to-low reproducibility was noted for specific fractional anisotropy, gray matter thickness, and white matter lesion counts.
• Physiological measurements showed variable reproducibility, from excellent for spectroscopy to low for regional blood flow.

Conclusions:

• Quantitative MRI/S measurements exhibit high longitudinal consistency in healthy subjects, defining inherent variability.
• Identifies specific sequences and regions requiring cautious interpretation due to lower reproducibility.
• Provides a foundation for interpreting MRI/S data in neurological disease assessment and treatment monitoring.