A decay-modeled compressed sensing reconstruction approach for non-Cartesian hyperpolarized ¹²⁹Xe MRI

Area of Science:

• Medical Imaging
• Magnetic Resonance Imaging
• Hyperpolarized Gas MRI

Background:

• Non-Cartesian acquisitions offer efficient k-space sampling in hyperpolarized 129Xe MRI.
• Existing gridded reconstructions struggle with magnetization decay artifacts and undersampling.

Purpose of the Study:

• To develop a compressed sensing (CS) reconstruction method for hyperpolarized 129Xe MRI.
• To incorporate magnetization decay processes into the CS forward model.
• To improve image sharpness, contrast, and undersampling capabilities.

Main Methods:

• Incorporated radio-frequency, T1, and decay processes into the forward model for iterative CS reconstruction.
• Validated the decay-modeled reconstruction using simulations.
• Tested the method in 2D/3D-spiral ventilation and 3D-radial gas-exchange MRI, comparing with gridded reconstructions.

Main Results:

• CS reconstructions showed a threefold increase in accuracy in simulations.
• In vivo imaging demonstrated improved sharpness and contrast for ventilation, with greater accuracy in undersampled data.
• Gas-exchange imaging, especially dissolved-phase, showed improved apparent SNR and discernable structure with CS.

Conclusions:

• A novel non-Cartesian CS reconstruction approach effectively models hyperpolarized 129Xe decay.
• This method enhances image quality, sharpness, and contrast, enabling up to threefold undersampling.
• The approach offers significant benefits for hyperpolarized gas MRI, improving accuracy and reducing scan times.