Fast MR parameter mapping using k-t principal component analysis.

Area of Science:

• Medical Imaging
• Biophysics
• Computational Science

Background:

• Magnetic resonance (MR) parameter quantification is crucial for diagnosing neurodegenerative diseases.
• Long scanning times currently limit the clinical adoption of MR parameter mapping.
• Accelerated data acquisition techniques are essential for routine clinical use.

Purpose of the Study:

• To adapt the k-t principal component analysis (PCA) method for accelerated MR parameter quantification.
• To evaluate the feasibility of significantly reducing scan times for inversion recovery fast imaging with steady state precession (TrueFISP) acquisitions.

Main Methods:

• Applied k-t PCA, a technique initially used for cardiac imaging, to MR parameter quantification.
• Undersampled k-t space and reconstructed aliased data using k-t Broad-use Linear Acquisition Speed-up Technique (BLAST) principles and temporal basis functions.
• Investigated the impact of varying basis functions and training data on reconstruction accuracy.

Main Results:

• Achieved up to an 8-fold acceleration in MR parameter quantification measurements.
• Enabled estimation of relaxation times T(1) and T(2) and spin density in a single slice.
• Acquired data with sub-millimeter in-plane resolution in just 6 seconds.

Conclusions:

• k-t PCA effectively accelerates MR parameter mapping, reducing acquisition times to clinically acceptable levels.
• This accelerated method holds promise for routine clinical application in neurodegenerative disease detection.
• The technique facilitates rapid, high-resolution quantification of key MR parameters.