| Purpose:（1）Based on Auto-prescription technology,to explore the effects of individualized low tube voltage scanning on the measurement of liver fat content and image quality in QCT.（2）Through dose matching,to investigate the impact of individualized low tube current scanning on the measurement of liver fat content and image quality.Materials and methods:（1）We prospectively collected 231 patients who received combined unenhanced CT of chest and upper abdomen in our hospital.The liver in the overlapping part of two scans in the same case was studied,in which the chest used individualized low-dose scanning（group A）and the upper abdomen used routine 120k Vp scanning（group B）.Group A selected patients scanned with 80k Vp and 100k Vp based on Auto-prescription technology,denoting them as group A1 and group A2.Group B was routine abdominal CT scan with 120k Vp,divided into groups B1 and B2,which corresponded to group A1 and A2.The remaining parameters of scanning and reconstruction were kept as the same for both groups.Post-adaptive statistical iterative reconstruction V（ASIR-V）was performed for group A from 40%to 80%with 20%intervals and 40%for group B.Images were transmitted to AW4.6 and QCT Pro workstations.On all groups of images,the CT values and SD values of the left lobe,right anterior lobe and right posterior lobe of liver in the overlapping scan of the chest and upper abdomen were measured on the two workstations.On the same image,the signal-to-noise ratio（SNR）and contrast-to-noise ratio（CNR）were calculated by using vertical spinal muscles as background noise.Bland-Altman analysis was applied to assess the consistency between the two measurements of liver fat content,and Pearson correlation analysis was used to construct the linear regression equation.The image quality was evaluated by two observers using the quintuple method and Kappa consistency test was performed to assess the inter-reader agreement.Hepatic fat percentage determined by QCT(Fat%QCT)and correlation,CT value,SD value,SNR,CNR and subjective image score were compared between group A and group B.（2）From February 2022 to December 2022,a total of 253 patients who received combined chest and upper abdominal CT scan in our hospital were prospectively collected.The overlapped liver in the two scans was studied,and 120k Vp individualized low-dose scan was used in the chest（group C）.The radiation dose is equivalent to the80k Vp and 100k Vp groups based on Auto-prescription technology,denoting as group C1and C2.The upper abdomen is scanned by routine 120k Vp（group D）,divided into group D1 and D2,which corresponded to group C1 and C2.Post-ASIR-V was performed for group A from 40%to 80%with 20%intervals and 40%for group B.Images were transmitted to AW4.6 and QCT Pro workstations.On all groups of images,the CT values and SD values of the left lobe,right anterior lobe and right posterior lobe of liver in the overlapping scan of the chest and upper abdomen were measured on the two workstations.On the same image,the SNR and CNR were calculated by using vertical spinal muscles as background noise.Paired sample t test was used to compare the difference between two measurements,and Bland-Altman analysis was applied to assess the consistency between the two measurements of liver fat content.The image quality was evaluated by two observers using the quintuple method and Kappa consistency test was performed to assess the inter-reader agreement.Hepatic fat percentage determined by QCT(Fat%QCT)and correlation,CT value,SD value,SNR,CNR and subjective image score were compared between group C and group D.Results:（1）The CT value and Fat%QCT of liver in group A were higher than those in group B（t=20.128-20.429,P<0.05）.When 40%post-ASIR-V is applied in group A,the SNR and CNR of group A1 and A2 were lower than those of group B1 and B2 with conventional tube voltage.When combining 60%/80%post-ASIR-V,the SNR and CNR of group A1 and A2 were higher than those of group B1 and B2 with conventional tube voltage combined with 40%post-ASIR-V,and the differences were all statistically significant（P<0.05）.There was no significant difference in Fat%QCT of ASIR-V images with different weights in A1 and A2 groups（F=0.004,0.002,P>0.05）,but there were significant differences in Fat%QCT between A1 and B1,A2 and B2 groups（t=21.916-23.601,P<0.05）.Under 40%post-ASIR-V,the Fat%QCT measured by A1 and B1,A2and B2 had a high consistency,showing a significant positive correlation（r=0.9397,0.9872,P<0.05）.Based on the liver fat content in group B,the linear equation of low k Vp was constructed.The linear regression equations under 80k Vp and 100k Vp were established as follows:y=0.976x+3.119 and y=1.007x+2.041,respectively（FIG.2,FIG.3）.The subjective scores of group A and group B had strong consistency（Kappa=0.755～0.914）.The highest subjective score was obtained in 60%post-ASIR-V of group A,with no statistically significant difference compared to group B（P>0.05）.In comparison with virtual routine 120k Vp,the radiation dose decreased by 28.10%and 14.20%for80k Vp and 100k Vp,respectively.（2）There was no significant difference in liver CT values between groups C and D（t=0.656～1.793,P>0.05）.With the increase of ASIR-V weight,the SD value of group C gradually decreased（P<0.05）,whereas the SNR and CNR gradually increased（P<0.05）.The SNR and CNR of group C1 and C2 with low tube current,reconstructed with40%post-ASIR-V,were lower than those of group D1 and D2 combined with conventional tube voltage and reconstructed with 40%ASIR-V.The SNR and CNR of group C1 and C2 combined with 60%/80%post-ASIR-V were higher than those of D1and D2 combined with 40%post-ASIR-V,and there were statistically significant differences（P<0.05）.There was no significant difference in Fat%QCT of ASIR-V images with different weights between groups C1 and C2（F=0.999,0.998,P>0.05）.The Fat%QCT values of groups C and D had a high consistency（r=0.9993,0.9995,P>0.05）.The subjective score of group C and group D showed a good consistency（Kappa=0.773～0.918）,which demonstrated a trend of increasing first and then decreasing.The highest subjective score was achieved in group C using 60%post-ASIR-V,and with no statistically significant difference compared to group D.For CT chest,no significant difference was found between the recommended radiation dose based on Auto-prescription technology and the actual radiation dose（t=1.105～1.760,P>0.05）.Conclusion:（1）There was a significant positive correlation between80k Vp/100k Vp and the measured liver fat content at 120k Vp QCT.The regression equation could be used to quantitatively analyze the liver fat content at low tube voltage.Low dose individualized scanning protocol for chest combined with 60%post-ASIR-V achieves a win-win goal of image quality and radiation dose.（2）For the low dose chest scan using Auto-prescription technology,the same radiation dose can be obtained by adjusting noise index（NI）with 120k Vp.In addition to the chest scan using individualized low tube current combined with 60%post-ASIR-V,the liver fat measurement can be accurately obtained and the image quality can be guaranteed.It provides the possibility to carry out the broad research of QCT to measure liver fat content and establish QCT as the reference standard to diagnose fatty liver. |
PDF Full Text Request