Optimization Of Imaging Quality And Evaluation Of Coronary Stenosis By Using Gemstone Spectral Imaging:an In Vitro Study

Part ⅠOptimization of monochromatic spectral imaging quality by using CT coronary angiography:A Phantom StudyObjective:To evaluate the different energy levels of virtual monochromatic spectral (VMS) coronary artery image qualities at 60bpm heart rate in order to determine the optimal VMS image levels of CT coronary angiography. Materials and Methods: Totally 3 different coronary artery models (internal diameter:3.00 mm) with different severities degree of stenosis (residual lumen diameter ratio:75%、50%、25%) were placed in a pulsating cardiac phantom (ALPHA 1-VT PC, Fuyo Corporation, Japan). The coronary phantom was scanned in 2 models:gemstone spectral imaging (GSI) and axial scan model (A-CTCA). All the spectral imaging data were analyzed by using GSI viewer to reconstruct the VMS images (40-140keV by 10 keV interval). The image Noise, signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) were calculated for each data set. For all groups of VMS image and A-CTCA, the imaging Noise, SNR, CNR of all coronary artery models were compared by using One-Way ANOVA and Tukey post hoc test. P<0.05 was considered as significant difference. Results:(1) There were significant differences of N among 40～140 keV of VMS and A-CTCA images (F=13.768, P<0.001) Noise values (≥60 keV) were significant higher than other keV intervals(<60 keV), while the Noise values (≥100 keV) were also significant higher than A-CTCA images (P<0.05). (2) For all the different severities of stenosis, there were significant difference of SNR and CNR among VMS and A-CTCA images (SNR:F =12.660, P<0.001; CNR:F=11.704, P<0.001;). From 60 to70 keV, SNR and CNR reached the highest value comparing to other keV intervals (P<0.05 or P>0.05). SNR and CNR from 60 to70 keV were significant higher than A-CTCA (P=0.000). Conclusion:VMS images from 60 to 70 keV levels, which significantly improved the coronary image quality by raising the SNR and CNR, and reducing the Noise in all severities of stenosis, may be the optimized keV choice for coronary stenosis diagnosis in the clinics.Part IIEvaluation of coronary stenosis by using the gemstone spectral CTCA:A Phantom StudyObjective:To evaluate diagnosis accuracy for coronary stenosis from diameter measurements of different severities of stenosis by using gemstone spectral imaging (GSI). Materials and Methods:Totally 3 different coronary artery models(internal diameter:3.00 mm) with different severities of stenosis (residual lumen diameter ratio:75%,50%,25%,) were placed in a pulsating cardiac phantom (ALPHA 1-VT PC, Fuyo Corporation, Japan). The coronary phantom was scanned in 2 models: gemstone spectral imaging (GSI) and axial scan model (A-CTCA). All the spectral imaging data were analyzed by using GSI viewer to reconstruct the VMS images (40-140keV by 10 keV interval) and A-CTCA. CT values (HU) of the remained lumen from all different stenositc lumens were measured. The residual rate (%) was calculated by dividing the diameters of the remained lumen to normal lumen on the reformatted short axial images. One-Way ANOVA was used to compare the residual rate measurement difference between VMS. Student t test was used to compare the residual rate measurement difference between VMS and A-CTCA. Bland-Altman test was used to compare the measured residual rates and the gold standard.P<0.05was considered as significant differences. Results:There were no significant differences among the VMS diameter measurements (P>0.05). So 70keV VMS with higher CNR and SNR was selected to compare with A-CTCA. From student t test, only significant difference between 70keV VMS and A-CTCA was found on 50% stenosis (t=4.617, P<0.042). From the Bland-Altman test, residual rate measurement from 70keV VMS was more accurate than A-CTCA by taking the real model stenosis rate as gold standard(t=14.560, P<0.000). In the severe severity of stenosis (25% residual lumen), GSI and A-CTCA both underestimated the degree of stenosis, however, in the moderate severity of stenosis (50% residual lumen), two scan models both overestimated the degree of stenosis. In the mild severity of stenosis (75%of the residual lumen), GSI slightly underestimated the degree of stenosis, while A-CTCA slightly overestimated the stenosis. The accuracy of the GSI measurement was slightly higher than that of A-CTCA.Conclusion:VMS image (70keV) showed more accurate than A-CTC A images on both diameter measurement and evaluation of the coronary stenosis rate.