〔Abstract〕 Abstract Objective To evaluate the clinical value of combining low-dose energy spectrum CT with virtual un-en- hanced (VUE) scanning and deep-learning image reconstruction (DLIR) in aortic CT angiography (CTA) . Meth- ods In a prospective study , 94 patients scheduled for aortic CTA were randomized into two groups : a low-dose en- ergy spectrum group and a standard 100 kVp enhancement group , with 47 patients in each . All patients initially underwent a true un-enhanced (TUE) scan at 120 kVp using adaptive statistical iterative reconstruction-V (ASIR- V) at 40% for image reconstruction . The low-dose group received enhanced scans using gemstone spectral imaging (GSI) mode with DLIR-H , producing 60 keV virtual monoenergetic images (VMIs) and VUE images . The stand- ard group was scanned at 100 kVp , with images reconstructed using ASIR-V at 50% . Parameters were measured including CT values , noise (SD) , signal-to-noise ratio (SNR) , and contrast-to-noise ratio (CNR) for key vascu- lar and muscular areas , alongside the effective radiation dose (ED) . Two radiologists evaluated the image quality using a 5-point scale . Results The low-dose group exhibited significantly higher SNR and CNR values in the as- cending aorta , descending aorta , abdominal aorta , and common iliac artery compared to the standard group (P < 0. 05) , with comparable subjective quality scores . The VUE images also demonstrated superior SNR values in the abdominal aorta , common iliac artery , and psoas major muscle , and CNR value in the ascending aorta compared to TUE images , with similar subjective quality. Importantly , the ED in the low-dose group was about 40% lower than that of the standard group . Conclusion Low-dose energy spectrum CT with DLIR in aortic CTA can significantly enhance SNR and CNR , while approximating the image quality of traditional TUE scans , thereby substantially re- ducing radiation exposure .