Comparison of Absolute Apparent Diffusion Coefficient (ADC) Values in ADC Maps Generated Across Different...

Area of Science:

• Radiology and Medical Imaging
• Biomedical Engineering
• Quantitative MRI

Background:

• Apparent diffusion coefficient (ADC) maps are vital for quantitative MRI analysis.
• Current ADC map generation relies on vendor-specific software, limiting standardization.
• Open-source software offers alternatives for image postprocessing and parametric map generation.

Purpose of the Study:

• To assess the reproducibility of absolute ADC values derived from different postprocessing software.
• To compare ADC values generated by in-line vendor software versus external postprocessing tools.

Main Methods:

• Diffusion-weighted (DW) images were acquired using a 1.5-T MRI scanner with three b values.
• ADC maps were generated using both the MRI vendor's in-line software and separate postprocessing software.
• Comparison involved regions of interest (ROIs) analyzed with statistical tests and graphical methods (Bland-Altman, Passing-Bablok).

Main Results:

• A statistically significant difference was observed in mean ADC values across different postprocessing software packages.
• Variability in ADC values was evident even when using identical baseline trace DW images.

Conclusions:

• The choice of postprocessing software significantly impacts absolute ADC values.
• Standardization of postprocessing algorithms is essential for reliable ADC quantification in histology.
• This standardization is particularly important for vendor-neutral archiving and cross-platform data analysis.