The Clinical Value Of3.0T Multiparametric Magnetic Resonance Imaging In Prostate Cancer Diagnosis

PartI Application of diffusion kurtosis imaging in the diagnosis ofprostate cancerObjective: To assess the value of the diffusional kurtosis imaging（DKI） fordistinguishing benign regions of prostate from malignant, and to find out its optimalthreshold for diagnosis. Material and method:41patients with prostate cancer and15withprostate hyperplasia confirmed by pathology, with age52-82, were involved in this study.All the patients were undertook conventional MRI and multi-b-value DWI with a3.0T MRscanner. Those b values were set between0to2500s/mm2. The image acquisition and dataanalysis were conducted by two skilled radiologists, and a total of127valid sets of DKIdata were statistic analysed, including58prostate cancer foci,41non-cancer tissue in thecentral lobe and28non-cancer tissue in the peripheral zone. Then the D and the Kparameter diagram of the prostate cancer tissue, the non-cancer tissue in the central lobeand in the peripheral zone of the prostate were measured by the DKI model formula. Thevariance of the D/K value in those different tissues was compared, the relationship betweenD/K and the Gleason score in prostate cancer were assessed; and the ROC curves wereanalysed in order to calculate the sensitivity and specificity of D and Kin the diagnosis ofprostate cancer.Result: There is a significant difference in D/K between the prostate cancer tissue and thenon-cancer tissue of the central lobe（Z=7.8672,P<0.0001;Z=-7.8086, P<0.0001）, thenon-cancer tissue of both the central lobe and peripheral zone（Z=3.6113,P=0.0003; Z=-4.7618, P<0.0001）, and prostate cancer tissue and non-cancer tissue of peripheral zone（Z=7.2792,P<0.0001; Z=-7.2505, P<0.0001）；The optimal threshold for D value was1.5432, while the optimal threshold for K value was0.7086, respectively; thus the samesensitivity and specificity in detecting a prostate cancer tissue were94.7%and95.5%. Thearea under the curve were0.980and0.977, respectively. However， D/K value is notrelevant with Gleason score（r=-0.023，P=0.891；r=-0.067，P=0.686）. Conclusion:TheDKI model is capable in reflecting the variation between the prostate cancer andnon-cancer tissues through D/K value, which plays a significant role in the diagnosis ofprostate cancer. PartII3.0TMR dynamic enhanced MRI in clinical diagnosis ofprostate cancerObject: to assess the diagnostic value of DCE-MRI in prostate cancer with thequantitative analysis of prostate cancer by DCE-MRI. Material and method:49patientswith prostate diseases were selected, aged between43and81, including32patients withprostate cancer and the other17with prostate hyperplasia. All the patients were undergoneconventional MRI and DCE-MRI scan, and then measure and compare the variation thevalue of Ktrans、Kep、Vebetween the prostate cancer and normal tissues. And under theROC curve analysis, we calculate the sensitivity and specificity ofKtrans、Kepand Vein thediagnosis of prostate cancer, as well as the correlation between Ktrans、Kep、Veand theGleason score in prostate cancer. Result: the value of both Ktrans、Kepare significantlydifferent between the prostate and normal tissues（P<0.001）, while Veis not（P>0.05）.The areas of Ktrans、Kepunder the ROC curve are the largest, and the sensitivity andspecificity of both parameters are92.5%、71.4%，90.6%、71.4%,yet, Veis clinicallyuseless in the diagnosis of prostate. Meanwhile Ktrans、Ve value is not relevant withGleason score，Kephas a little relevance with Gleason score. Conclusion: DCE-MRIquantitative analysis plays a significant role in the diagnosis of prostate cancer, and mayhelp to distinguish prostate tumors。 PartIII The clinical value of magnetic resonance spectroscopicimaging in the diagnosis of prostate cancerObjective: To discuss magnetic resonance spectroscopic imaging (MRSI) in prostatecancer screening applications, and to find out its optimal threshold for diagnosis. Methods:30patients with prostate cancer，31with prostate hyperplasia and one with neoplasiaconfirmed by pathology, were involved in this study. MRS choline (Cho), creatine (Cr),citrate (Cit) values of each voxel were measured，with (Cho+Cr)/Cit ratio to weight theratio of prostate tissue metabolism substance.The results:There was a significantdifference in (Cho+Cr)/Cit ratio of positive and negative voxel in central leaf andperipheral zone,the difference has a statistically significant (P <0.001). The area under thecurve were0.938and0.993for central leaf and peripheral zone,the optimal threshold for (Cho+Cr)／Cit ratio in central leaf was0.805, the sensitivity and specificity in detecting aprostate cancer tissue were83%and89.8%. while the optimal threshold for(Cho+Cr)／Cit ratio in peripheral zone was0.885, the sensitivity and specificity in detecting a prostatecancer tissue were98.2%and93.8%. Conclusion: MRS has a high value in the diagnosisof prostate, and the diagnostic value in the peripheral zone was higher than the centrallobe.