A novel approach for accelerating mouse abdominal MRI by combining respiratory gating and compressed sensing

作者：Farias Alexandre Rodrigues; Medeiros Daniel de Castro; Magalhaes Hermes Aguiar; Dutra Moraes Marcio Flavio; Mendes Eduardo M A M^*

来源：Magnetic Resonance Imaging, 2018, 50: 45-53.

DOI：10.1016/j.mri.2018.03.005

• Compressed sensing
• Respiratory gating
• Undersampling pattern
• Small animal
• Magnetic resonance imaging (MRI)
• Variable density

Purpose: To combine the technique of respiratory gating and compressed sensing (CS) with the objective of accelerating mouse abdominal magnetic resonance imaging (MRI).
Materials and methods: To obtain the maximum acceleration, phase-encoding data from a phantom and mouse were obtained on a 4.7 Tesla scanner using the respiratory gating technique. The fully sampled data (FSD) were used to construct reference images and to provide samples to simulate retrospective undersampled data (UD) acquisition using respiratory gating. The UD and 95% of the UD on acceleration 2-5 rates were acquired and used for image reconstruction by CS. Quantitative assessment of reconstructed images was performed by structural similarity index (SSIM), peak signal-to-noise ratio (PSNR) and root mean square error (RMSE).
Results: The proposed method can accelerate phantom and mouse abdominal MRI acquisition between 2 and 4 rates by reducing the amount of FSD. For phantom UD acquisition, the mean time was reduced in 45.9% and for the acquisition of 95% of UD in 67.8%. For mouse abdominal image UD acquisition, the mean time was reduced in 44.6% and for the acquisition of 95% of UD in 62.5%. The metrics results show that the reconstructed image from UD and 95% of UD by using CS maintains an optimal agreement with their reference images (similarity above 0.88 for phantom and 0.93 for mouse).
Conclusion: This study presents a novel approach to accelerate mouse abdominal MRI combining respiratory gating technique and CS without the use of expensive hardware and capable of achieving up to 4 acceleration rate without image degradation.

• 出版日期2018-7