MRI Reconstruction Using Markov Random Field and Total Variation as Composite Prior

Area of Science:

• Medical Imaging
• Signal Processing
• Computational Science

Background:

• Magnetic resonance imaging (MRI) reconstruction quality is enhanced by integrating prior knowledge of coefficient statistical dependencies.
• Markov Random Field (MRF) models show superior performance in capturing intraband dependencies for MRI reconstruction compared to inter-scale models.

Purpose of the Study:

• To develop a novel MRI reconstruction method incorporating a composite prior.
• To enhance reconstruction accuracy by utilizing an anisotropic MRF model and Total Variation (TV).

Main Methods:

• A composite prior combining an anisotropic MRF model and Total Variation (TV) was developed.
• A data-driven method for adaptive estimation of MRF parameters was proposed.
• A Bayesian framework was used to define a position-dependent regularization and derive a compact reconstruction algorithm with a novel soft-thresholding rule.

Main Results:

• The proposed method demonstrates superior performance in MRI reconstruction.
• Experimental results confirm the effectiveness of the novel composite prior and reconstruction algorithm.
• The method outperforms existing state-of-the-art techniques in the field.

Conclusions:

• The developed MRI reconstruction method effectively leverages composite priors for improved image quality.
• The novel adaptive parameter estimation and position-dependent regularization contribute to enhanced reconstruction accuracy.
• This approach represents a significant advancement in MRI reconstruction techniques.