Partial fourier reconstruction through data fitting and convolution in k-space.

Area of Science:

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

Background:

• Partial Fourier acquisition accelerates MRI scans but faces challenges with phase information and artifacts.
• Existing methods often fail to improve phase information and struggle with rapid phase changes, leading to image distortions.

Purpose of the Study:

• To develop a novel and robust method for partial Fourier reconstruction.
• To address limitations in phase information estimation and artifact reduction in existing techniques.

Main Methods:

• Introduced a k-space convolution approach for partial Fourier reconstruction.
• Implicitly estimated phase information in k-space via data fitting.
• Recovered unacquired k-space data using Hermitian operation and k-space convolution.

Main Results:

• The proposed method consistently yielded the lowest error levels in spin echo and gradient echo imaging.
• Significant improvements were observed in images with rapid phase changes.
• Reconstructed images showed lower error in both magnitude and phase maps compared to conventional methods.

Conclusions:

• The k-space convolution method offers superior performance for partial Fourier reconstruction.
• This technique effectively enhances phase accuracy and reduces artifacts, particularly in challenging imaging conditions.
• The method provides improved image quality for fast MRI applications.