Pharmacokinetic modeling of dynamic contrast-enhanced MRI using a reference region and input function tail

Area of Science:

• Medical Imaging
• Biophysics
• Pharmacokinetics

Background:

• Quantitative analysis of dynamic contrast-enhanced MRI (DCE-MRI) is crucial for disease assessment.
• Measuring the arterial input function (AIF) in DCE-MRI is technically challenging and often limits accuracy.
• Existing models require precise AIF measurements, hindering widespread clinical application.

Purpose of the Study:

• To introduce and validate the Reference Region and Input Function Tail (RRIFT) approach for quantitative DCE-MRI.
• To assess RRIFT's performance in simulations and in vivo studies, particularly under suboptimal temporal resolutions and noise.
• To compare RRIFT's accuracy and precision against the established Extended Tofts Model (ETM).

Main Methods:

• RRIFT utilizes a reference tissue and the washout portion of the AIF to derive kinetic parameters.
• Simulations were conducted with 100 parameter combinations across various temporal resolutions (5-30s) and noise levels.
• In vivo validation involved 8 glioblastoma multiforme patient studies, comparing RRIFT with ETM using both measured and literature-based AIF tails.

Main Results:

• RRIFT estimated kinetic parameters ($$, $$) with median errors within 20% in simulations.
• RRIFT demonstrated superior accuracy and precision over ETM at temporal resolutions slower than 10s.
• In vivo, RRIFT showed excellent agreement with ETM (CCC > 0.90 for $$ and $$) using measured AIF tails.

Conclusions:

• The RRIFT method enables quantitative DCE-MRI analysis by providing absolute kinetic parameters.
• This approach is robust and viable for both simulations and in vivo applications, even with temporal resolutions as low as 30s.
• RRIFT offers a practical solution for overcoming AIF measurement difficulties in DCE-MRI.