An AI tool measuring the pericoronary fat attenuation index agreed closely with conventional workstation analysis across five CT scanners, a retrospective study of 1,125 patients found. It achieved 84.98% accuracy and an AUC of 0.851, with 94.84% of measurements within limits of agreement. Researchers say automation could standardise coronary inflammation assessment, though prospective multicentre studies are needed.
AI Tool Matches Workstation FAI Measurement Across CT Scanners
Source: EMJ