Font Size: a A A

The Quantitative Evaluation Of MR Brain Perfusion Maps Generated By Different Commercial Software

Posted on:2014-04-25Degree:MasterType:Thesis
Country:ChinaCandidate:L FuFull Text:PDF
GTID:2284330434472851Subject:Imaging and nuclear medicine
Abstract/Summary:PDF Full Text Request
Part I:The quantitative evaluation of ROI in MR brain perfusion maps generated by different commercial softwarePurpose:To explore the application of MR brain perfusion generated by different commercial software.Materials and Methods:Forty-five patients underwent MR T2/T2*-weighted dynamic susceptibility contrast MRI, DSC-MR1. The MR perfusion imaging data were postprocessed by using three different commercial software packages:Syngo Verio3.0, UIH SW001.001, Perfusion Mismatch Analyzer (PMA) Ver.3.4.0.6. The same ROI were adopted by each software. And the value of cerebral blood flow (CBF), cerebral blood volume (CBV), mean transit time (MTT), and Time to Peak(TTP)were compared with each other. The results of rCBF, rCBV, rMTT, rTTP from different software were alos compared by using t-test, Pearson correlation, intraclass correlation coefficient and Bland-Altman analysis.Results:The difference of every absolute parameter value (CBF, CBV. MTT, TTP) obtained respectively from perfusion software packages had statistical significance. As for the relative parameters value (rCBF, rCBV, rMTT, rTTP), no significant differences were found using paired t-test among them (P>0.05). The Pearson correlation analysis indicated they all significantly correlated with each other (r>0.95). The ICC value is also high(ICC>0.95). And A strong correlation was also found in the Bland-Altman analysis.Conclusion:Absolute parameter values obtained from different software were significant different. The evaluation of relative parameter values can eliminate the difference of results caused by different post-processing algorithm, and can be used preferably in qualitative and quantitative analysis. Part I:The similarity analysis of MR brain perfusion maps generated by different commercial softwarePurpose:To explore the similarity of MR brain perfusion generated by different commercial software.Materials and Methods:Ten patients underwent MR T2/T2*-weighted dynamic susceptibility contrast MRI, DSC-MRI. The MR perfusion imaging data were postprocessed by using three different commercial software packages:Syngo Verio 3.0, UIH SW001.001, Perfusion Mismatch Analyzer (PMA) Ver.3.4.0.6. Compared the maps of cerebral blood flow (CBF), cerebral blood volume (CBV), mean transit time (MTT), and Time to Peak(TTP) with each other by using the normalized correlation with Student’s t-test.Results:A validation of MR brain perfusion algorithm is performed by comparing the computed perfusion parameters among different vendors, i.e Syngo, AW and UIH. For parameter CBF, the normalized correlation (NC) test results between UIH and PMA, UIH and Syngo are both good and that between PMA and Syngo is a fair. For CBV, the NC test between UIH and PMA, UIH and Syngo are both just fair and that between PMA and Syngo is poor. For MTT, the NC test result between UIH and PMA is a fair, that between UIH and Syngo is good and that between PMA and Syngo is poor. For TTP, all three tests are fair and particularly the similarity between UIH and Syngo is significant higher than the other two.Conclusion:The perfusion result from UIH is similar to the results both from Syngo and PMA even though the results between Syngo and PMA are sometimes not quite close.
Keywords/Search Tags:MR brain perfusion, ROI, Deconvolution, quantitative analysisMR brain pertusion, similarity normalized correlation
PDF Full Text Request
Related items