Super-Resolution Deep-Learning Reconstruction Boosts CTA Stent Assessment Accuracy in Multicenter Trial

Super-resolution deep-learning reconstruction (SR-DLR) improved in-stent restenosis detection on coronary CTA: per-stent accuracy 89.1% vs 79.1% for hybrid iterative reconstruction (HIR) in a reader, with 73 patients and 110 stents, along with better image quality.

• Objective image quality: SR-DLR increased CNRstent (44.0±20.1 vs 33.6±15.7), reduced SAIR (0.44±0.36 vs 0.71±0.55), and improved edge sharpness (469±261 vs 221±130 HU/mm) (all p<.05).
• Subjective image quality and diagnostic confidence both improved from median 3 to 4 (on 5-point scale) with SR-DLR (both p<.001).
• Accuracy benefits held in stents ≤3.0 mm diameter, with per-stent accuracy up to 93.2% vs 81.4%.

Automated summary

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