An aliasing artifacts reducing approach with random undersampling for spatiotemporally encoded single-shot MRI

Area of Science:

• Magnetic Resonance Imaging (MRI)
• Image Reconstruction
• Medical Physics

Background:

• Echo planar imaging (EPI) is susceptible to field inhomogeneity.
• Spatiotemporally encoded (SPEN) MRI offers enhanced field inhomogeneity immunity.
• SPEN MRI requires effective reconstruction methods to address undersampling artifacts.

Purpose of the Study:

• To develop and evaluate a hybrid reconstruction scheme for SPEN MRI.
• To reduce aliasing artifacts caused by vast undersampling in SPEN images.
• To improve the overall image quality of SPEN MRI.

Main Methods:

• A hybrid scheme combining random sampling, singular value decomposition (SVD), and compressed sensing (CS).
• Implementation of super-resolved reconstruction for SPEN imaging.
• Validation through numerical simulations, water phantom experiments, and in vivo rat brain imaging.

Main Results:

• The hybrid scheme significantly reduced aliasing artifacts in reconstructed SPEN images.
• Comparable spatial and temporal resolutions were maintained post-reconstruction.
• Demonstrated efficiency in reducing artifacts under vast undersampling conditions.

Conclusions:

• The proposed hybrid reconstruction scheme is effective for SPEN MRI, especially in undersampled scenarios.
• This method enhances the practical applicability of SPEN MRI by improving image quality.
• The technique offers a valuable solution for mitigating artifacts in advanced MRI sequences.