Progressive magnetic resonance image reconstruction based on iterative solution of a sparse linear system

Area of Science:

• Computed Imaging
• Signal Processing
• Medical Imaging

Background:

• Image reconstruction from nonuniformly sampled spatial frequency domain data is a significant challenge in computed imaging.
• Existing reconstruction techniques often have limitations in their modeling and implementation.

Purpose of the Study:

• To present a novel, efficient iterative method for image reconstruction.
• To address limitations of current reconstruction techniques.

Main Methods:

• Solving a system of linear equations using an iterative approach.
• Applying an orthogonal transformation to create a sparse system matrix.
• Utilizing the conjugate gradient method for solving the transformed system.

Main Results:

• Successfully reconstructed images from numerical phantom and magnetic resonance imaging (MRI) spiral data.
• Demonstrated that the computational load is comparable to standard gridding methods.
• Validated the practical utility of the proposed reconstruction method.

Conclusions:

• The developed iterative method offers an efficient and practical solution for image reconstruction from nonuniformly sampled data.
• The orthogonal transformation and conjugate gradient approach effectively handle dense system matrices.
• This technique shows promise for applications in computed imaging, including MRI.