5.0 T versus 1.5 T and 3.0 T upper abdominal MRI: a prospective two‑cohort study of image quality and diagnostic confidence

OBJECTIVES: To compare 5.0 T magnetic resonance imaging (MRI) image quality versus 1.5 T and 3.0 T in patients and healthy volunteers, so as to validate its clinical feasibility. MATERIALS AND METHODS: This prospective study included a patient cohort (348 consecutive patients undergoing upper abdominal MRI) and a healthy volunteer cohort (31 volunteers scanned on all three scanners). Two radiologists independently assessed image quality subjectively (5-point Likert scale: anatomical detail, artifacts, overall image quality, diagnostic confidence) and measured signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR). Subjective and objective metrics were compared across field strengths using overall and pairwise comparisons. RESULTS: Objective metrics (left/right/mean liver SNR, muscle/spleen SNR, liver-spleen CNR) showed a significant gradient of 5.0 T > 3.0 T > 1.5 T in both cohorts (all p < 0.001). Subjectively, compared with 1.5 T and 3.0 T, 5.0 T received the lowest scores for cardiac motion artifact and liver signal uniformity on T2-weighted imaging (T2WI) and showed greater distortion and ghosting artifacts on diffusion-weighted imaging (DWI) (all p < 0.05). Nevertheless, 5.0 T was significantly superior in anatomical detail, SNR, and overall image quality (all p < 0.05). In patients, 5.0 T also achieved the highest diagnostic confidence (p < 0.05). Results from healthy volunteers were largely consistent with those from the patient cohort. CONCLUSION: 5.0 T upper abdominal MRI, with a primary focus on liver imaging, is clinically feasible, offering superior anatomical detail, SNR, overall image quality, and diagnostic confidence over 1.5 T and 3.0 T, though cardiac motion artifacts, liver signal inhomogeneity on T2WI, and geometric distortion on DWI require further optimization. KEY POINTS: Question Is 5.0-T MRI superior to 1.5 T and 3.0 T for liver imaging, and does its diagnostic benefit differ between diseased and non-diseased livers? Findings 5.0 T provided significantly higher SNR, image quality, and diagnostic confidence than 1.5 T and 3.0 T, despite increased cardiac motion and liver signal inhomogeneity. Clinical relevance 5.0-T MRI significantly improves liver image quality and diagnostic confidence over 3.0 T and 1.5 T, supporting its clinical utility despite increased artifacts that require further optimization.