Interpolated compressed sensing for 2D multiple slice fast MR imaging.

Area of Science:

• Medical Imaging
• Biophysics
• Signal Processing

Background:

• Sparse Magnetic Resonance Imaging (MRI) accelerates scans by undersampling k-space data.
• Increased undersampling in sparse MRI compromises image quality, notably reducing contrast-to-noise ratio (CNR).

Purpose of the Study:

• To introduce an interpolated Compressed Sensing (iCS) method for multi-slice 2D sparse MRI.
• To improve imaging speed and reduce data size without sacrificing image quality or CNR.

Main Methods:

• The iCS method leverages k-space data from adjacent slices during acquisition.
• Missing k-space data is estimated using neighboring slice data and a weighting function derived from low-resolution reference images.
• In-vivo human foot MRI was used to evaluate the iCS method's performance.

Main Results:

• The iCS reconstruction method demonstrated reduced average image error compared to conventional sparse MRI.
• The average CNR was improved using the iCS method at identical undersampling rates.
• Feasibility and performance of iCS were confirmed in human subjects.

Conclusions:

• The proposed iCS method effectively enhances imaging speed and data reduction in sparse MRI.
• iCS maintains or improves image quality and CNR, addressing key limitations of conventional sparse MRI.
• This technique offers a promising solution for efficient multi-slice 2D sparse MRI in clinical settings.