Parallel imaging and compressed sensing combined framework for accelerating high-resolution diffusion tensor imaging...

Area of Science:

• Medical Imaging
• Biophysics
• Signal Processing

Background:

• High-resolution diffusion tensor imaging (DTI) faces challenges with low signal-to-noise ratio and reconstruction artifacts, hindering efficient acquisition.
• Parallel imaging (PI) and compressed sensing (CS) are techniques used to accelerate MRI scans but require careful implementation for high-resolution DTI.

Purpose of the Study:

• To develop and evaluate a novel framework combining parallel imaging (PI) and compressed sensing (CS) for accelerated high-resolution DTI acquisition.
• To address the limitations of low signal-to-noise ratio and sensitivity to artifacts in high-resolution DTI.
• To improve acquisition efficiency while maintaining image quality and accuracy of diffusion tensor parameters.

Main Methods:

• Proposed a new method named anisotropic sparsity SPIRiT, integrating PI and CS with motion error correction and PI calibration.
• Employed inter-image correlation of diffusion-weighted images through anisotropic signals to improve sparsity.
• Utilized a multishot variable density spiral DTI acquisition scheme for demonstration.

Main Results:

• The proposed anisotropic sparsity SPIRiT method demonstrated superior performance compared to CG-SENSE, CS-based joint reconstruction, and other PI-CS combined methods.
• Qualitative and quantitative analyses showed better preserved image quality and more accurate DTI parameters with the proposed method.
• Experiments were conducted at acceleration factors ranging from 3 to 5 in brain DTI.

Conclusions:

• The anisotropic sparsity SPIRiT framework effectively accelerates high-resolution DTI acquisition.
• The method leverages sharable information across different diffusion encoding directions to enhance efficiency.
• This approach offers a promising solution for improving the practicality of high-resolution DTI.