Deep learning VIBE reconstruction boosts upper abdominal MRI quality in gynecologic cohort

Deep learning–reconstructed volumetric interpolated breath-hold examination (VIBE) improved image quality, reduced artifacts, and reduced noise versus standard VIBE in 60 women, especially without contrast; signal-to-noise and contrast-to-noise ratios did not differ (p≥0.35).

• Prospective study of 60 women; three radiologists rated four axial sequence types (standard and deep learning VIBE, non-contrast and contrast-enhanced) on standardized 4-point Likert scales.
• Interobserver agreement was moderate for image quality and artifacts but low for noise; quantitative signal-to-noise and contrast-to-noise ratios showed no significant difference (all p≥0.35).

Automated summary

RadPigeon summaries are original and for information only. They are not clinical advice.