80-kVp triple rule out CT with deep learning reconstruction reduces radiation and contrast while preserving image quality

80-kVp triple-rule-out CT with deep learning image reconstruction reduced radiation dose by 35.1% (7.29 vs 11.22 mSv) and contrast volume by 18.8% (51 vs 62 mL) while maintaining image quality and diagnostic performance.

• Prospective randomized study of 159 patients comparing 80-kVp triple-rule-out CT with deep learning image reconstruction (DLIR) versus 100-kVp ASIR-V; consensus of two chief radiologists served as reference standard for coronary stenosis.
• Compared with 100-kVp ASIR-V, the DLIR group had lower image noise, higher signal-to-noise and contrast-to-noise ratios, comparable or superior edge rise slope, and higher figure-of-merit across nine artery sites.
• AI-reader and junior radiologists performed similarly in classifying coronary stenosis in both groups, with benefits maintained across BMI strata (<25 and ≥25 kg/m2).

Automated summary

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