Evaluation of Apparent Diffusion Coefficient Repeatability and Reproducibility for Preclinical MRIs Using...

Area of Science:

• Preclinical Imaging
• Magnetic Resonance Imaging (MRI)
• Quantitative Imaging

Background:

• Co-clinical trials require reliable preclinical MRI data for translation to human studies.
• Assessing the performance of apparent diffusion coefficient (ADC) measurements in preclinical MRI is crucial for standardization.

Purpose of the Study:

• To evaluate the repeatability, reproducibility, and bias of ADC measurements in preclinical MRI systems.
• To compare the performance of preclinical MRI scanners to clinical MRI standards using standardized protocols.

Main Methods:

• Multi-institutional study involving 10 preclinical MRI scanners from 3 vendors across 6 field strengths.
• Utilized a temperature-controlled phantom for absolute reference and spatial uniformity assessment.
• Acquired diffusion-weighted imaging (DWI) data over multiple days, comparing centralized and site-based analysis.

Main Results:

• Excellent short-term (intra-exam) and long-term (multiday) repeatability with within-system coefficient of variance (wCV) of 0.73% and 1.26%, respectively.
• Cross-system reproducibility coefficient (RDC) was 0.188 µm²/ms (17% relative to reference standard).
• Most systems (8/10) exhibited low absolute bias (<4%); two high-bias scanners impacted overall RDC. Site-specific analysis introduced >2% additional variance for 2/10 systems.

Conclusions:

• Well-calibrated preclinical MRI systems demonstrate high repeatability and reproducibility for ADC measurements.
• Performance metrics are comparable to clinical MRI, scaled for bore size, supporting co-clinical trial applications.
• Standardized procedures are essential for minimizing variance and ensuring data integrity in preclinical quantitative MRI.