Single breath-hold lung MRI with AI compressed sensing detects solid nodules with 98% sensitivity for ≥4 mm vs. photon-counting CT

Lung MRI with AI-compressed sensing in single breath-hold achieved 98.1% sensitivity (95% CI 90.2-99.7) for solid nodules ≥4 mm vs PCD-CT, with excellent size agreement (CCC 0.985). Missed calcified nodules <4 mm. Prospective single-center study.

• Prospective single-center diagnostic accuracy study in 97 healthy adults (67% men, mean age 58; 34% smokers) with same-day PCD-CT reference standard.
• Overall accuracy for solid nodules: 87.3%, rising to 97.7% for nodules ≥4 mm; Lung-RADS 4A/4B showed complete agreement; mean size underestimation by MRI: 1.02 mm.
• Limitation: single-center design; missed calcified nodules <4 mm (51.2% of undetected nodules); not externally validated.

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