| Part 1 Feasibility study of Revolution CT low kilovolt ultra-fast reverse flow scan in CTA imaging of abdominal aortaand lowe extremity arteriesObjective: To explore the feasibility of Revolution CT low kilovolt ultra-fast reverse flow scan in CTA imaging of abdominal aorta and lower extremity arteries.Materials and Methods: A total of 95 patients who underwent CTA examination of abdominal aorta and lower extremity arteries in Revolution CT from September2020 to March 2021 were randomly divided into two groups: group A(n = 47)and group B(n = 48).In group A,the tube voltage was 80 Kv,the tube speed was 0.28s/r,and the pitch was 0.992 ROI.The scanning trigger ROI was set at the distal calf artery on the bilateral ankle joint plane to scan in the direction of reverse blood flow from the sole of the foot to the abdomen,while in group B,the tube voltage was 100 Kv,the speed of the tube was 1s/r,and the pitch was 0.984 ROI.The scanning trigger ROI was set in the distal abdominal aorta,and the scanning direction was from the abdomen to the sole of the foot.The contrast medium with iodine concentration of 400 mg I/ml was used in both groups.Multi-phase injection was used in group An and conventional single-phase injection was used in group B.After the scanning is completed,the image is transmitted to the post-processing workstation,and MPR,MIP and VR are used to reconstruct and reconstruct the image.The levels of proximal abdominal aorta,distal abdominal aorta,left and right common iliac artery,left and right femoral artery(upper segment),left and right superficial femoral artery(middle segment),left and right poplitealartery,left and right tibiofibular trunk,left and right distal leg artery(anterior tibial artery or posterior tibial artery)were measured and calculated,and the average CT value,signal-to-noise ratio(SNR),contrast-to-noise ratio(CNR)and noise(SD)of corresponding vessels and muscles were calculated.At the same time,the two diagnostic physicians used a four-point method to subjectively score the image quality.DLP,CTDI,ED,scan length,exposure time,vascular CT value,SD value,SNR,CNR and subjective score were evaluated in group An and group B.P < 0.05 showed that there was a statistical difference between the two groups.Result:The average acquisition exposure time in group A was(4.4 ±0.2)s,and the average acquisition exposure time in group B was(32.4 ±2.2)s.DLP(321.6 ±29.4m Gy * cm),CTDI(2.6 ±0.2m Gy)and ED(1.48 ±0.1m Sv)in group A were significantly lower than those in group B(DLP(745.3 ±212.7m Gy*cm),CTDI(5.8 ±1.4m Gy)and ED(4.32 ±1.2m Sv)(P <0.05).In the part of CT value,the CT values of proximal abdominal aorta,distal abdominal aorta and common iliac artery in group A were significantly lower than those in group B(P < 0.05).The CT values of common femoral artery and superficial femoral artery were similar between the two groups(P > 0.05),and the CT values of popliteal artery,tibiofibular trunk artery and distal leg artery in group A were significantly higher than those in group B(P < 0.05).In the part of SNR,the SNR of proximal abdominal aorta,distal abdominal aorta,common iliac artery and common femoral artery in group A was lower than that in group B(P<0.05),while the SNR in superficial femoral artery and distal leg artery in group A was significantly higher than that in group B(P<0.05),but there was no significant difference in SNR in popliteal artery and tibiofibular artery between the two groups(P>0.05).In the part of CNR,the CNR values of proximal abdominal aorta,distal abdominal aorta,common iliac artery and common femoral artery in group A were lower than those in group B(P<0.05),but there was no significantdifference in CNR values of superficial femoral artery,popliteal artery,tibiofibular trunk artery and distal leg artery between the two groups(P>0.05).The subjective score of abdominal aorta-iliac artery in group A was lower than that in group B(P < 0.05),and there was no significant difference between the two groups in femoral-popliteal artery,calf artery and foot artery.Conclusion: CTA scanning of abdominal aorta and lower extremity arteries with Revolution CT low kilovolt ultra-fast reverse blood flow scan is significantly lower than that of conventional methods,and the image quality meets the diagnostic requirements,which i clinically feasible.Part 2 Noise reduction assessment by deep learning based image reconstruction algorithm in CTA of abdominal aorta and lower extremity arteriesObjective:The effect of image reconstruction(deep learning image reconstruction,DLIR)and adaptive statistical iterative reconstruction(adaptive statistical iterative reconstruction,ASIR)on noise reduction of abdominal aorta and lower extremity arteries in CTA imaging of abdominal aorta and lower extremity arteries.Materials and Methods:The original data of 47 patients who underwent CTA examination of abdominal aorta and lower extremity arteries were retrospectively included.The images of 0.625 mm were reconstructed by mixed ASIR-V reconstruction(ASIR-V20%,ASIR-V50%,ASIR-V80%)and deep learning of high,middle and low modes(DLIR-H,DLIR-M,DLIR-L).The regions of interest(ROI)were delineated in the proximal abdominal aorta,distal abdominal aorta,left and right common iliac artery,left and right femoral artery(upper segment),left and right superficial femoral artery(middle segment)and left and right popliteal artery,respectively,and the CT value,SD value,signal-to-noise ratio(SNR)and contrast-to-noise ratio(CNR)of the region of interest were measured and evaluated objectively.The blood vessels of the lower extremities were divided into four segments(abdominal aorta-common iliac artery,common femoral-popliteal artery,calf artery and foot artery)to subjectively evaluate the granularity and sharpness of the image.Result:With the increase of ASIR-V reconstruction weight(ASIR-V20% <ASIR-V50% < ASIR-V80%)and DLIR reconstruction intensity(DLIR-L <DLIR-M < DLIR-H),the SD value of each measurement layer decreased gradually,while the SNR value and CNR value increased gradually.Among the six reconstruction methods,DLIR-H reconstruction and ASIR-V80%reconstruction have lower SD values and higher SNR and CNR values.The SDvalues of DLIR-H images were significantly lower than those of ASIR-V20% and ASIR-V50%(P < 0.001),while the SNR and CNR values of DLIR-H images were significantly higher than those of ASIR-V20% and ASIR-V50% images(P <0.001).There was no significant difference in SD,SNR and CNR values between DLIR-H images and ASIR-V80% images(P > 0.05).Through the comparison of the subjective scores of the six groups,it was found that the granularity score of DLIR-H reconstructed images was the best in the whole course of abdominal aorta-foot artery(mean 1.8-2.0 points).The sharpness score of DLIR-H reconstruction was significantly higher than that of ASIR-V80% in the whole course of abdominal aorta and foot artery(mean 2.93.3).The results showed that the image definition of DLIR-H group was significantly better than that of ASIR-V80% group,and the overall quality of DLIR-H image was the best.Conclusion: among all the reconstruction algorithms,the image reconstructed based on deep learning model can significantly reduce the noise of CTA images of abdominal aorta and lower extremity arteries and improve the image quality.Among the three deep learning reconstruction algorithms of DLIR(H,M,L),DLIR-H has the best denoising efficiency and is a balanced image with image noise and sharpness. |
PDF Full Text Request