Artificial intelligence iterative reconstruction in lower extremity computed tomography angiography (CTA) of diabetic patients: Improved visualization of distal and collateral arteries.

This article focuses on evaluating the performance of artificial intelligence iterative reconstruction (AIIR), a novel deep learning (DL)-based computed tomography (CT) reconstruction algorithm, in enhancing the visualization of distal and collateral arteries in lower extremity CT angiography (CTA) of diabetic patients. In a retrospective study of 59 patients with type II diabetes, AIIR demonstrated significantly improved visualization scores for most distal arteries and collateral circulation compared to routine hybrid iterative reconstruction (HIR), alongside reduced image noise and higher signal-to-noise and contrast-to-noise ratios. These improvements suggest that AIIR may facilitate more accurate assessment of peripheral arterial disease in diabetic limbs, potentially aiding clinical decision-making for diabetic foot ulcer management. The study acknowledges limitations including its focus on routine dose settings and the lack of comparison with other vendors' algorithms.