AI图像后处理技术对于低电压低对比剂冠状动脉CT血管成像图像质量的影响研究OA

Objective To explore the influence of artificial intelligence(AI)-based post-processing technology on the image quality in double low(80 kVp tube voltage,40 ml contrast agent)coronary computed tomography angiography(CCTA).Methods This retrospective study enrolled patients who underwent CCTA examinations at Hai'an Hospital of Traditional Chinese Medicine between January 2021 and October 2023,with a body mass index(BMI)≤25 kg/m2.Patients were randomly assigned to either the 100 kVp group or the 80 kVp group using a random number table.Images were reconstructed using both conventional post-processing methods and AI algorithms,and subjective and objective image quality assessments of the coronary arteries were performed.Differences in general characteristics,subjective and objective image quality assessments,and radiation dose between groups were compared.Results A total of 102 patients were enrolled in this study,including 53 patients in the 100 kVp group and 49 patients in the 80 kVp group.The 100 kVp group demonstrated significantly higher average coronary artery CT values[(527.00±80.00)HU vs.(343.00±60.00)HU,P<0.001],image noise[(26.00±7.00)HU vs.(21.00±7.00)HU,P=0.002)],and signal-to-noise ratio(SNR)[(20.30±4.20)vs.(16.30±4.30),P<0.001]compared to the 80 kVp group.However,there was no statistically significant difference in contrast-to-noise ratio(CNR)between the two groups[(24.50±4.00)vs.(22.60±6.20),P=0.099].There was no statistically significant difference in subjective image quality scores between images processed using artificial intelligence(AI)and traditional post-processing methods.The radiation dose in the 80 kVp group was significantly lower than that in the 100 kVp group.Conclusions Double low CCTA can reduce the radiation dose for patients,and AI-reconstructed images enhance physician workflow efficiency while maintaining coronary artery image quality.