Résumé
Objective To investigate w hether the iterative reconstruction (iDose 4 ) technique improves imaging quality of the low-radiation-dose w hole brain CT perfusion (CTP). Methods Thirty-five consecutive patients w ith clinical y suspected ischemic stroke w ere col ected. Bril iance 256 iCT w as used to perform low-radiation-dose w hole brain CTP, and the filtered back projection (FBP) and iDose 4 algorithm w ere used to conduct image reconstruction. The noise and signal to noise ratio of the 2 kinds of reconstruction algorithms, as w el as the imaging quality of each parameter map w ere compared. Results The effective dose of the w hole brain CTP w as 2.2 mSv. Compared w ith FBP, the noise of each region of interest in the iDose4 Tmax map was decreased significantly ( P<0.05) and the signal to noise ratio was increased significantly (P<0.05). The imaging quality scores (median, interquartile range) reconstructed by FPB group w ere significantly low er than by iDose 4 for cerebral blood flow (CBF) map ( 5.00 [3.00-6.00]vs. 6.00 [5.00-6.00]; Z= -2.784, P=0.005), cerebral blood volume (CBV) map ( 6.00 [5.00-6.00] vs. 6.00 [6.00-7.00]; Z= -3.674, P<0.001), and mean transit time (MTT) map (4.00 [3.00-5.00] vs. 5.00 [4.00-6.00]; Z=3.394, P=0.001). The proportions of the poor quality in CBF map ( 34.3%vs. 11.4%;χ2 =7.036, P=0.030), CBV map (11.4%vs.2.9%; χ2 =7.485, P=0.024 ) and MTT map (28.6%vs.11.4%;χ2 =5.318, P=0.070) reconstructed by FBP w ere significantly higher than by iDose 4 . Conclusions The iDose4 technique may improve imaging quality of low er-radiation-dose CTP.
Résumé
Objective To compare the consistency and correlation of multiple breath-hold (BH) with respiratory-triggered (RT) 1H-MRS for quantification of hepatic lipid content. Methods Sixty subjects were underwent RT 1H-MRS of the liver (Couinaud segment VII) and BH 1H-MRS at 1.5 Tesla Magnetic Resonace Imaging (MRI). The peak areas of water and methylene obtained on RT and BH 1H-MRS were recorded respectively and the liver fat fraction was calculated. Pearson correlation coefficient , Bland-Altman 95% limit of agreement, and concordance correlation coefficient were calculated. Results Mean liver fat fraction measured in RT and BH 1H-MRS were (8.6 ± 8.7)% and (9.4 ± 9.3)% respectively. There was a strong correlation between RT and BH 1H-MRS(r = 0.973, P < 0.000 1, concordance correlation coefficient = 0.95). With the Bland-Altman method, 91.7% data points were within the 95% limits of agreement. Conclusion RT and BH 1H-MRS are alternative tools for intrahepatic lipid quantification. These two methods have a strong correlation and perfect consistency.
Résumé
<p><b>OBJECTIVE</b>To investigate the changes in whole brain gray matter volume in patients with trigeminal neuralgia using voxel-based morphometry (VBM).</p><p><b>METHODS</b>Twenty-eight patients with trigeminal neuralgia and 28 healthy controls underwent magnetic resonance imaging with a Philips 1.5T MRI scanner. VBM was used to compare the structural differences in the whole brain gray matter between the two groups based on the DARTEL after data preprocessing with SPM8 software package.</p><p><b>RESULTS</b>Compared with the healthy controls, the patients with trigeminal neuralgia presented with decreased gray matter volume in several brain regions including the bilateral middle temporal gyrus, bilateral superior/middle frontal gyrus, left pre-/post-central gyrus, right fusiform and anterior cingulate gyrus.</p><p><b>CONCLUSIONS</b>Patients with trigeminal neuralgia had abnormal gray matter volume in some brain regions associated with perception and processing of pain sensation. These changes may provide clues for further exploration of the neuropathogenic basis of trigeminal neuralgia.</p>