Exploiting similarity in adjacent slices for compressed sensing MRI

Area of Science:

• Medical Imaging
• Biophysics

Background:

• Magnetic Resonance Imaging (MRI) fundamental characteristics limit scan speedup.
• Compressed Sensing (CS) MRI techniques use image sparsity in transform domains for faster acquisition.
• MRI's poor signal-to-noise ratio (SNR) is linked to image slice thickness.

Purpose of the Study:

• To develop a novel Compressed Sensing (CS) MRI method for rapid acquisition.
• To improve the signal-to-noise ratio (SNR) in MRI reconstructions.
• To enhance the overall image reconstruction quality in MRI.

Main Methods:

• Utilizing the difference between adjacent slices as the sparse domain for CS MRI.
• Acquiring thick MRI slices instead of thin ones.
• Reconstructing thin slices from thick slice data by leveraging inter-slice similarity.

Main Results:

• The proposed method improves the total SNR of the reconstructed image.
• Reconstruction quality of the proposed method outperforms existing CS MRI techniques.
• The method achieves superior results with the same number of measurements.

Conclusions:

• Acquiring thick slices and reconstructing thin slices offers improved SNR and image quality in CS MRI.
• This approach addresses limitations of current CS MRI methods.
• The novel sparse domain provides a more effective strategy for rapid and high-quality MRI.