Parallel magnetic resonance imaging acceleration with a hybrid sensing approach

Area of Science:

• Medical Imaging
• Biophysics
• Computer Vision

Background:

• Magnetic Resonance Imaging (MRI) acquisition is time-consuming, leading to patient discomfort and image artifacts.
• Compressed Sensing (CS) and parallel MRI (pMRI) are techniques to reduce MRI scan times without sacrificing image quality.
• Combining CS and pMRI offers synergistic benefits for accelerated MRI acquisition and improved image reconstruction.

Purpose of the Study:

• To propose a novel method combining a combined CS approach with pMRI for enhanced acceleration of MRI acquisition.
• To investigate a hybrid undersampling strategy in K-space, integrating regular and random sampling patterns.
• To evaluate the efficacy of the proposed method using established image quality metrics.

Main Methods:

• A combined Compressed Sensing (CS) method is integrated with parallel Magnetic Resonance Imaging (pMRI).
• K-space data undersampling is performed using a combination of regular and traditional random sampling techniques.
• Image reconstruction is achieved through nonlinear conjugate gradient optimization.

Main Results:

• The proposed method demonstrates significant improvements in image reconstruction compared to traditional schemes.
• Average reconstruction error, Universal Image Quality Index (Q-index), and Peak Signal-to-Noise Ratio (PSNR) showed remarkable enhancements of up to 59%, 63%, and 39%, respectively.
• Numerical simulations validate the effectiveness of the combined CS and pMRI approach.

Conclusions:

• A simple, efficient, and computationally less intensive method for MRI reconstruction is presented by combining CS and pMRI.
• This approach offers substantial improvements in image reconstruction quality and acquisition speed.
• The findings have significant implications for reducing MRI imaging times and enhancing patient experience.