| Part I Evaluation of imaging quality of high b-value RESOLVE at 1.5 Tesla MR for prostate cancerObjective By comparison of RESOLVE and conventional single shot echo(Single-shot echo planner imaging,SS-EPI)at 1.5T magnetic resonance imaging system with conventional b-value or high b-value,to evaluate the image quality of DWI with RESOLVE for prostate cancer(PCa).Methods a total of 83 consecutive patients who underwent MRI examination for clinical suspective prostate lesions were enrolled between May 2016 and February 2017.32 patients met the inclusion criterion were finally included in the study.Inclusion criterion:(1)Underwent MRI examination of RESOLVE DWI and SS-EPI DWI with conventional b-value and high b-value.(2)Without history of prostate tumor.Without surgery,radiotherapy,chemotherapy or endocrine therapy for PCa.(3)Patients had not received a prostate biopsy within 8 weeks before MRI examination.(4)PCa was confirmed by using ultrasound guided prostate biopsy within 6 weeks after MRI examination.Exclusion criterion:(1)Patients can not tolerate MRI examination or image quality was not enough for evaluation.(2)With no evidence for PCa or ROI cannot be setted accurately.(3)The pathological results were not consistent with the MRI image.Average age of patients was 72.3±7.6 years old.PSA were 0.65-5673.9ng/ml.Conventional T2 WI,SS-EPI and RESOLVE were performed at SIEMENS MAGNETOM Aera 1.5T MR.Conventional b-value(0,800s/mm2)and high b-value(0,1600s/mm2)were used in both SS-EPI and RESOLVE.DWI sequences were divided into four groups:(1)RESOLVE,b-value:0,1600s/mm2.(2)RESOLVE,b-value:0,800s/mm2.(3)SS-EPI,bvalue:0,1600s/mm2.(4)SS-EPI,b-value:0,800s/mm2.By using 5-point Likert scale,image quality of image distortion,anatomical detail resolution,lesion conspicuity,signal-to-noise ratio(SNR),lesion contrast noise ratio(CNR)and contrast were assessed by two radiologists independently.Kappa test and Bowker test were used to evaluate the consistency of the two readers.Differences between groups were analyzed using Wilcoxon test for paired sample.P<0.05 was considered as statistically significant.Results At b=0,1600s/mm2,the ? value for image distortion,anatomical detail resolution and lesion conspicuity were 0.84,0.618 and 0.805,respectively.At b=0,800s/mm2,the ? value for image distortion,anatomical detail resolution and lesion conspicuity were 0.584,0.72 and 0.587,respectively.P values of kappa test were <0.001.P values of Bowker test were>0.05.These resultsindicated the consistency of readers was good.Objective image quality index: Group(1)(RESOLVE,b-value:0,1600s/mm2): SNR,CNR and contrast were 23.83±5.21,14.66±4.72 and 2.73±0.87,respectively.Group(2)(RESOLVE,b-value:0,800s/mm2): SNR,CNR and contrast were 40.40±7.27,14.78±5.31 and 1.62±0.26,respectively.Group(3)(SS-EPI,b-value:0,1600s/mm2): SNR,CNR and contrast were 78.48±11.28,35.76±11.91 and 1.81±0.30,respectively.Group(4)(SS-EPI,b-value : 0,800s/mm2):SNR,CNR and contrast were 98.8±11.09,28.84±11.63 and 1.36±0.19,respectively.The SNR and CNR of SS-EPI were superior to that of RESOLVE at both b-value.With high b-value,the contrast of RESOLVE and SS-EPI-DWI were both improved compared with conventional b-value.With high b-value,the RESOLVE achieved the best contrast.Conclusion RESOLVE provides better image quality.When using high b-value,although the SNR was reduced,the lesions contrast of RESOLVE was significantly improved compared to RESOLVE with conventional b value and to SS-EPI.Part II The valve of high b-value RESOLVE at 1.5 Tesla in the diagnosis for prostate cancerObjective By comparison of RESOLVE with conventional b-value and high b-value,to investigate the value of RESOLVE with high b-value at 1.5T in the diagnosis for prostate cancer.Methods The clinic and medical information of 78 patients underwent RESOLVE-DWI examination were retrospectively analyzed.Of 78 patients,32 were prostate cancer cases.The average age of cancer patients was 72.3±7.6 years old,and the PSA was 0.65-5673.9ng/ml.46 were benign prostatic hyperplasia(BPH)cases.The average age of BPH patients was 68.6±7.3,and the PSA was 0.59-58.59ng/ml.Inclusion criterion:(1)Underwent MRI examination of RESOLVE DWI with conventional b-value(0,800s/mm2)and high b-value(0,1600s/mm2).(2)Without history of prostate tumor.Without surgery,radiotherapy,chemotherapy or endocrine therapy for PCa.(3)Patients had not received a prostate biopsy within 8 weeks before MRI examination.(4)PCa or BPH were confirmed by using ultrasound guided prostate biopsy within 6 weeks after MRI examination.Gleason scores were assessing for PCa patients.Exclusion criterion:(1)Patients can not tolerate MRI examination.(2)image quality was not enough for evaluation.(3)With no evidence for PCa or ROI can not be setted accurately.(4)The pathological results were not consistent with the MRI image.Conventional T1 WI,T2WI,and RESOLVE-DWI with conventional b-value(0,800s/mm2)and high b-value(0,1600s/mm2)were performed at SIEMENS MAGNETOM Aera 1.5T superconducting magnetic resonance scanner.Prostate were divided into he basal,intermediate,and apical parts of the prostate on left or right side.The basal and intermediate parts on each side were subdivided into peripheral zone and central gland.The detail information of samples obtained by each needle were recorded.DWI Images were interpreted based on PI-RADS v2(prostate imaging reporting and data system v2)by two radiologists.Samples with score 4 or 5 were considered as PCa.Any disagreement in assessment was resolved by discussion.The pathological results of Twelve-core ultrasound guided prostate systematic biopsy were used as reference standard.The sensitivity,specificity of RESOLVE conventional b-value and high b value in the diagnosis of prostate cancer were calculated.According to PCa zone confirmed by pathological results and the high signal area on DWI image.ROI were drawed on ADC map of conventional b-value and high b-value.ROI with the same shape and size were drawed on the corresponding part in the noncancerous regions of the other side.In addition,the ADC of PCa and benign prostatic hyperplasia(BPH)were compared.The ADC ratio of cancer area to contralateral corresponding noncancerous regions was calculated.The correlation between ADC value and of Gleason risk grades was also analyzed.The difference of ADC between cancer area and corresponding noncancerous area were analyzed by using t-test for matched samples.The difference of ADC between PCa and BPH were analyzed by using Wilcoxon test for paired sample.The Correlation between ADC value and Gleason score using Spearman rank correlation test.Results At b=0,800s/mm2,the sensitivity,specificity were 75.0%,82.6%,respectively.At b=0,1600s/mm2,the sensitivity,specificity were 93.8%,93.5%,respectively.The ADC ratio of high b-value(0.614±0.122)was lower than that of conventional b-value(0.682±0.144).At b=0,1600s/mm2,the ADC of PCa and BPH were(0.81±0.13)x10-3 mm2/S and(1.21±0.14)x10-3mm2/S,respectively.At b=0,800s/mm2,the ADC of PCa and BPH were(1.00±0.15)x10-3 mm2/S and(1.48±0.20)x10-3mm2/S,respectively.At b=0,800s/mm2,the ADC value was negatively correlated with the Gleason score(r=-0.376,p=0.049).At b=0,1600s/mm2,the negative correlation between ADC value and Gleason score was improved(r=-0.423,p=0.025).Conclusion The sensitivity,specificity of high b-value RESOLVE-DWI at 1.5T MR are superior to that of conventional value.The ADC ratio of high b-value is lower than that of conventional b-value.ADC of PCa is lower than that of BPH.These results indicate that high b-value RESOLVE-DWI at 1.5T MRis valuable tool for the diagnosis of prostate cancer.However,the sample size of the current study is small,the results should be further validated in multi-center studies with large sample size. |
PDF Full Text Request