Vendor-independent AI denoising boosts low-dose chest CT image quality versus iterative reconstruction

Image-domain deep learning denoising significantly improved objective and subjective image quality of low-dose chest CT compared to iterative reconstruction (all p<0.05), while enabling ~76% dose reduction vs standard-dose CT. Equivalence could not be confirmed due to retrospect…

• The low-dose protocol delivered ~76% lower effective radiation dose compared with the standard-dose protocol.
• Subjective image quality of lung parenchyma and mediastinal soft tissue was significantly better with denoising than with low-dose SAFIRE (5-point scale, all p<0.05).
• Retrospective, single-center design comparing different patient groups prevented equivalence or non-inferiority conclusions versus standard-dose CT.

Automated summary

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