Accelerated 3D mnoulti-echo spin-echo sequence with a subspace constrained reconstruction for whole mouse brain T 2...

Area of Science:

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

Background:

• Quantitative mapping is crucial for preclinical research.
• Current methods are time-consuming, limiting whole-brain coverage.
• Accelerated imaging techniques are needed for efficient whole-brain analysis.

Purpose of the Study:

• To accelerate whole-brain quantitative mapping in preclinical settings.
• To develop and validate an advanced reconstruction method for faster MRI acquisition.
• To enable efficient monitoring of biological processes like metastasis growth.

Main Methods:

• Employed a highly undersampled 3D multi-echo spin echo sequence with variable density Poisson sampling.
• Utilized advanced iterative reconstruction based on linear subspace constraints.
• Compared different subspaces (EPG simulations, calibration images) and investigated subspace dimension for quantitative mapping precision.

Main Results:

• Achieved robust quantitative mapping estimations using an EPG-based dictionary.
• Determined a subspace dimension of 6 as optimal for balancing quantitative precision and image quality.
• Enabled whole-brain quantitative mapping and detection/monitoring of small metastases (<500 μm) with high acceleration factors.

Conclusions:

• Subspace-based reconstruction is effective for 3D quantitative mapping.
• The method allows acceleration factors up to 8, reducing acquisition time to 25 minutes for whole-brain mouse imaging.
• This technique is suitable for monitoring metastasis growth in preclinical models.