Accelerated Brain DCE-MRI Using Iterative Reconstruction With Total Generalized Variation Penalty for Quantitative...

Area of Science:

• Medical Imaging
• Radiology
• Biophysics

Background:

• Dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) is crucial for brain tumor assessment.
• Quantitative pharmacokinetic analysis provides vital information but often requires long acquisition times.
• Accelerated imaging techniques are needed to improve clinical workflow and patient comfort.

Purpose of the Study:

• To evaluate the feasibility of using undersampled k-space data with iterative reconstruction for quantitative pharmacokinetic analysis in brain DCE-MRI.
• To assess the accuracy of pharmacokinetic parameters derived from accelerated acquisition compared to fully sampled data.

Main Methods:

• Retrospective analysis of eight brain DCE-MRI scans.
• Simulated acceleration by a factor of 4 using two sparse k-space sampling strategies: golden ratio radial and Cartesian random sampling.
• Image reconstruction using an iterative method with total generalized variation (TGV) penalty.
• Voxel-wise pharmacokinetic analysis using the extended Tofts model (K^trans) and a 2-compartment exchange model (F_B, V_B).

Main Results:

• Pharmacokinetic parameter maps from undersampled data were comparable to those from fully sampled data.
• Average error in volume mean was approximately 3.54%, with most values below 5%.
• Average total relative error was ~0.115, and average cross-correlation was ~0.962, indicating high accuracy and similarity.

Conclusions:

• The TGV-based iterative reconstruction method accurately estimates DCE-MRI pharmacokinetic parameters from sparsely sampled k-space data.
• Accelerated acquisition (factor of 4) using this method is feasible for reliable clinical application in brain imaging.
• This approach can potentially reduce scan times without compromising quantitative diagnostic information.