ABSTRACT
Objective@#To investigate the value of texture analysis based on MR ADC map of prostate in differentiating between low-grade and high-grade prostate cancer (PCa).@*Methods@#PCa confirmed by pathology after radical prostatectomy were analyzed retrospectively, all patients underwent multiparametric MRI before radical prostatectomy, including T1WI,T2WI and DWI. On the ADC map, ROI was drawn manually to encompass the whole tumor by ITK-SNAP software. The python-based pyradiomics package was used to extract 105 texture features. The intraclass correlation coefficient was used to evaluate the repeatability of the texture features. The independent sample t test or Mann-Whitney U test was used to exclude features that had no significant difference between low grade and high grade PCa. Lasso regression model and 5 fold cross validation method were used to obtain texture feature combination of the highest performance and develop a classification modelfor discriminating low from high grade PCa. ROC curve was used to evaluate the diagnostic efficiency of the model.@*Result@#Ninety patients with PCa confirmed by pathology after radical prostatectomywere analyzed retrospectively,including 36 patients with low-level PCa (GS≤3+4) and 54 patients with high-level PCa (GS≥4+3). The area under curve of the model was 0.841, with sensitivity 69.6% and specificity 91.2%, which was significantly higher than single texture feature or traditional mean ADC value.@*Conclusion@#Texture analysis based on MRI-ADC map of prostate could be used to discriminate low grade PCa from high grade PCa.
ABSTRACT
Objective To investigate the value of texture analysis based on MR ADC map of prostate in differentiating between low?grade and high?grade prostate cancer (PCa). Methods PCa confirmed by pathology after radical prostatectomy were analyzed retrospectively, all patients underwent multiparametric MRI before radical prostatectomy, including T1WI,T2WI and DWI. On the ADC map, ROI was drawn manually to encompass the whole tumor by ITK?SNAP software. The python?based pyradiomics package was used to extract 105 texture features. The intraclass correlation coefficient was used to evaluate the repeatability of the texture features. The independent sample t test or Mann?Whitney U test was used to exclude features that had no significant difference between low grade and high grade PCa. Lasso regression model and 5 fold cross validation method were used to obtain texture feature combination of the highest performance and develop a classification modelfor discriminating low from high grade PCa. ROC curve was used to evaluate the diagnostic efficiency of the model. Result Ninety patients with PCa confirmed by pathology after radical prostatectomywere analyzed retrospectively,including 36 patients with low?level PCa (GS≤3+4) and 54 patients with high?level PCa (GS≥4+3). The area under curve of the model was 0.841, with sensitivity 69.6% and specificity 91.2%, which was significantly higher than single texture feature or traditional mean ADC value. Conclusion Texture analysis based on MRI?ADC map of prostate could be used to discriminate low grade PCa from high grade PCa.
ABSTRACT
Objective The purpose of this study was to compare the image quality of readout-segmented EPI (RS-EPI) and that of standard single-shot echo-planar imaging (SS-EPI) in patients with scrotal diseases. Methods Initial diagnosis of scrotal diseases were included in the prospective study, all patients underwent scrotum routine MRI and RS-EPI, SS-EPI sequence at 3.0 T. A total of 38 patients were recruited qualitative assessment ,and 29 patients proved by operation and pathology (malignant 21 cases and benign 8 cases) were included quantitative analysis.For qualitative comparison of image quality, two readers independently assessed the two sets of DWI, which consisted of identification of structure and geometric distortion of scrotum (epididymis,testes, lesion) using a 5-point Likert scale. For assessment of image parameters, signal-to-noise ratio (SNR), contrast, contrast-to-noise ratio (CNR),apparent diffusion coefficient value of normal testis and testis distortion ratio of anatomical(T2WI)/DWI fusion image were calculated. Inter-observer agreement was assessed with the Kappa statistics, the differences of scores of RS-EPI and SS-EPI were compared by using Wilcoxon rank-sum test, and image parameters were compared by using paired sample t test. Results There was good reader agreement in the scores, the Kappa value were 0.77, 0.74, 0.80, 0.87. The difference of identification of structure and geometric distortion on RS-EPI and SS-EPI had statistic significance, the RS-EPI was superior to SS-EPI in image quality (all P<0.05).The SNR, contrast of benign lesion for two sequences had no statistic significance(P values were 0.352, 0.124, respectively), but the difference of CNR had statistic significance(P<0.05). The SNR,contrast and CNR of malignant lesion for two sequences had statistic significance (all P<0.05). The mean size of the testis was (35.61 ± 9.78)mm, the mean distance in fusion image of RS-EPI or SS-EPI with T2WI were (3.80 ± 1.32)mm and (7.54 ± 2.62)mm, and the mean distortion ratios of the two set were (11.1 ± 0.6)% and (22.4 ± 19.2)%, respectively, the difference of two sets had statistic significance(P<0.05). There was no significant difference in the ADC obtained by using either DWI method(t=0.396,P=0.796), the mean ADC values of normal testes for RS-EPI and SS-EPI, respectively, were (1.18 ± 0.06) × 10-3mm2/s, (1.24 ± 0.11) × 10-3mm2/s.Conclusion RS-EPI DWI images is a feasible technique in the scrotum for producing high-resolution DWI with reduced geometric distortion and offers potentially superior image quality compared to SS-EPI at 3.0 T.
ABSTRACT
Objective To determine whether hybrid three dimensional diffusion tensor imaging (3D DTI) contributes incremental value to standard T2WI technique for assessing neurovascular bundles (NVB) around the prostate. Methods This retrospective institutional review board-approved study included 69 consecutive patients with prostate tumor who underwent MRI including conventional T2WI and 3D DTI . DTI data were postprocessed and hybrid 3D DTI and axial T2W images were obtained. Three radiologists with one, five and thirteen years of experience in reading prostate MRI and one urologist with three years of surgical experience in urology who were blinded to patient data independently recorded their levels of preference on a five-point scale of the NVBs around the prostate on the basis of T2WI alone and hybrid 3D DTI and T2WI, respectively. The differences of scores of T2WI and hybrid 3D DTI and T2WI of the four doctors were compared by using nonparametric Wilcoxon rank. Results The average scores of hybrid 3D DTI to T2WI and alone T2WI to assess NVBs for 3 radiologists and one urologist were 4.4±0.6, 4.3±0.8, 4.2± 0.6, 4.9±0.3 and 2.9±0.8, 3.0±1.1, 1.6±0.7, 3.8±0.5, respectively. The hybird 3D DTI to T2WI improved the discrimination abilities of NVBs around the prostate for 3 radiologists and one urologist (Z values were-12.791,-9.737,-14.538,-14.901, P contributes significant incremental value to the standard T2WI technique for assessing NVB around the prostate.