Free-breathing echo-planar imaging based diffusion-weighted magnetic resonance imaging of the liver with prospective acquisition correction

Objective: To evaluate soft tissue contrast and image quality of a respiratory-triggered echo-planar imaging based diffusion-weighted sequence (EPI-DWI) with different b values for magnetic resonance imaging (MRI) of the liver. Methods: Forty patients were examined. Quantitative and qualitative evaluation of contrast was performed. Severity of artifacts and overall image quality in comparison with a T2w turbo spin-echo (T2-TSE) sequence were scored. Results: The liver-spleen contrast was significantly higher (P < 0.05) for the EPI-DWI compared with the T2-TSE sequence (0.47 +/- 0.11 (b50); 0.48 +/- 0.13 (b300); 0.47 +/- 0.13 (b600) vs 0.38 +/- 0.11). Liver-lesion contrast strongly depends on the b value of the DWI sequence and decreased with higher b values (b50, 0.47 +/- 0.19; b300, 0.40 +/- 0.20; b600, 0.28 +/- 0.23). Severity of artifacts and overall image quality were comparable to the T2-TSE sequence when using a low b value (P > 0.05), artifacts increased and image quality decreased with higher b values (P < 0.05). Conclusion: Respiratory-triggered EPI-DWI of the liver is feasible because good image quality and favorable soft tissue contrast can be achieved.