Field-inhomogeneity-corrected low-rank filtering of magnetic resonance spectroscopic imaging data

Area of Science:

• Medical Imaging
• Spectroscopy
• Signal Processing

Background:

• Low signal-to-noise ratio (SNR) is a significant challenge in magnetic resonance spectroscopic imaging (MRSI).
• Existing low-rank approximation denoising methods leverage MRSI data's partial separability but are sensitive to field inhomogeneity.
• Field inhomogeneity can violate assumptions, degrading denoising performance in practical MRSI applications.

Purpose of the Study:

• To develop a robust field-inhomogeneity-corrected low-rank filtering method for MRSI.
• To improve denoising performance and SNR in practical MRSI data acquisition.

Main Methods:

• Proposed a novel low-rank filtering technique incorporating field inhomogeneity correction.
• Applied the method to process MRSI data, addressing deviations from partial separability.
• Utilized in vivo experiments to validate the denoising approach.

Main Results:

• The proposed method demonstrated effective denoising of MRSI data.
• Field inhomogeneity correction significantly improved the robustness of low-rank filtering.
• Enhanced SNR was achieved in practical MRSI scans.

Conclusions:

• The field-inhomogeneity-corrected low-rank filtering method offers a more reliable solution for MRSI denoising.
• This approach effectively addresses a key limitation of previous MRSI denoising techniques.
• The method shows promise for improving the quality and utility of in vivo MRSI data.