Effect of Visualized Respiratory Training on Patient Cooperation and Image Quality in Gadoxetic Acid Disodium‐Enhanced Liver <scp>MRI</scp> : A Randomized Controlled Trial

BACKGROUND: Respiratory motion artifacts impair diagnostic accuracy in gadoxetic acid disodium (Gd-EOB-DTPA)-enhanced liver MRI. Conventional verbal pre-scan breathing training often yields inconsistent patient cooperation and image quality, creating an urgent clinical need for optimized training methods. PURPOSE: To evaluate the impact of a visualized respiratory training method integrated into clinical workflow on patient cooperation, image quality, and examination efficiency for Gd-EOB-DTPA-enhanced liver MRI. STUDY TYPE: Prospective, single-center, randomized controlled trial. POPULATION: One hundred and eighty-four patients (135 males; mean age 53.5 ± 13.3 years) scheduled for first-time Gd-EOB-DTPA-enhanced liver MRI. FIELD STRENGTH/SEQUENCE: 3 T; Dynamic contrast-enhanced imaging was performed with a 3D fast spoiled gradient-echo sequence, covering pre-contrast, arterial, portal venous, transitional, and hepatobiliary phases. ASSESSMENT: Primary outcome was patient cooperation rate assessed by blinded MRI technologists. Secondary outcomes included motion artifact scores (rated by three blinded radiologists), Signal-to-Noise Ratio (SNR), Contrast-to-Noise Ratio (CNR), transient severe motion (TSM) incidence, and examination duration. STATISTICAL TESTS: Student's t-test, Mann-Whitney U test, Chi-square test, and Gamma correlation; two-tailed p < 0.05 indicated significance. RESULTS: The experimental group (85 patients) had a significantly higher full cooperation rate (56.5% [48/85] vs. 27.3% [27/99], OR = 3.46, 95% CI: 1.87, 6.40, γ = 0.406), significantly lower motion artifact scores in arterial/portal venous/transitional phases, and significantly higher arterial and transitional phase SNR/CNR. The experimental group also showed significantly shorter preparation time (14.40 ± 2.36 vs. 16.16 ± 2.20 min; mean difference: -1.76 min; 95% CI: -2.42, -1.101). Scanning time did not differ significantly between groups (24.60 ± 1.60 vs. 24.89 ± 2.21 min; p = 0.329). TSM incidence did not differ significantly between groups (1.18% vs. 5.05%, p = 0.219). DATA CONCLUSION: Visualized respiratory training significantly improves patient cooperation, image quality, and workflow efficiency in Gd-EOB-DTPA-enhanced liver MRI, with no additional resource requirements supporting its routine clinical adoption. EVIDENCE LEVEL: 2. TECHNICAL EFFICACY: Stage 1.