ObjectiveTo investigate the diagnostic value of CT-derived fractional flow reserve (CT-FFR) and fat attenuation index (FAI) based on artificial intelligence-assisted diagnostic software in coronary artery stenosis. MethodsA retrospective analysis was conducted on patients clinically suspected of coronary artery syndrome who underwent coronary computed tomography angiography at Guangdong Province Traditional Chinese and Western Medicine Hospital between June 2021 and May 2025. Patients were divided into two groups according to scanning protocols: group A underwent conventional retrospective electrocardiography-gated scanning, while group B used Flash_ChestPlin mode. Invasive coronary angiography data served as the gold standard for diagnosing vascular stenosis (stenosis rate<50% defined as negative group, ≥50% with clinical symptoms as positive group). Radiation dose was compared between the two scanning protocols. The diagnostic efficacy of CT-FFR, pericoronary FAI, and transluminal attenuation gradient (TAG) based on artificial intelligence system for coronary stenosis was analyzed, including sensitivity, specificity, and area under the curve (AUC). ResultsA total of 567 vessels from 189 patients were analyzed, including 105 males, 84 females with a mean age of (62.5±12.3) years and a mean body mass index of (24.21±3.5) kg/m2. There were 112 patients in the group A and 77 patients in the group B. The radiation dose in the group B was significantly lower than that in the group A [69.7 (58.1, 84.1) mGy·cm vs. 420.4 (338.6, 514.2) mGy·cm, P<0.001]. Significant differences in FAI and CT-FFR were observed between negative and positive groups under both scanning protocols (P<0.05), while no significant difference existed in TAG (P>0.05). In the group A, the AUC values for diagnosing stenosis were 0.925 for CT-FFR, 0.610 for FAI, and 0.516 for TAG. Corresponding values in the group B were 0.889, 0.677, and 0.548 respectively, with CT-FFR demonstrating optimal diagnostic performance. ConclusionUnder both conventional scanning and Flash scanning, the artificial intelligence-based CT-FFR demonstrates good diagnostic performance for coronary artery stenosis, and the Flash protocol significantly lowers radiation dose, indicating substantial potential for clinical application.
Objective To explore the application value of artificial intelligence (AI) pulmonary artery assisted diagnosis software for suspected pulmonary embolism patients. Methods The data of 199 patients who were clinically suspected of pulmonary embolism and underwent pulmonary artery CT angiography (CTA) from June 2016 to December 2021 were retrospectively analyzed. Images of pulmonary artery CTA diagnosed by radiologists with different experiences and judged by senior radiologists were compared with the analysis results of AI assisted diagnostic software for pulmonary artery CTA, to evaluate the diagnostic efficacy of this software and low, medium, and senior radiologists for pulmonary embolism. The agreement of pulmonary embolism based on pulmonary artery CTA between the AI software and radiologists with different experiences was evaluated using Kappa test. Results The agreement of the AI software and the evaluation of pulmonary embolism lesions by senior radiologists based on pulmonary artery CTA was high (Kappa=0.913, P<0.001), while the diagnostic results of pulmonary artery CTA AI software was good after judged by senior radiologists based on pulmonary artery CTA (Kappa=0.755, P<0.001). Conclusions The AI software based on pulmonary artery CTA diagnosis of pulmonary embolism has good consistency with diagnostic images of radilogists, and can save a lot of reconstruction and diagnostic time. It has the value of daily diagnosis work and worthy of clinical promotion.