| Identification value of IVIM related parameters and texture characteristics for prostate cancer and hyperplasiaObjectiveTo explore the value of IVIM related parameters and texture characteristics in identifying prostate cancer（PCa）and prostate hyperplasia（BPH）.MethodsPatients who underwent prostate MRI from August 2020 to December 2021 were collected and had definite pathology were collected.The final study included 43 patients,including 23 in the prostate cancer group and 20 in the prostate hyperplasia group.The IVIM and T2 WI sequence images were selected for the study.The area of interest was manually delineated on GE 4.6 postprocessing software MADC,averaged three times,obtain IVIM related parameter values D,D*,f,D value is water molecular diffusion coefficient（slow diffusion）,D* microcirculation diffusion coefficient（fast diffusion）and f（D / D + D*）,compare the differences between the two groups,draw the ROC curve,the area under the curve,evaluate the differential efficiency of the above parameters on prostate cancer and hyperplasia;The T2 WI sequence was opened with ITK-SNAP software,the ROI of each layer of the cancer focus was manually delineated,multiple texture features were extracted with texture analysis software,and screening modeling was performed,the ROC curve was drawn,the area under the curve was calculated,and the discriminatory efficiency of texture features was evaluated.Results1.Among the IVIM correlation parameters,the intergroup analysis showed the measured P values of D,D* and f of 0.001,0.002 and 0.000,and their AUC values of 0.827,0.216 and 0.982,respectively.2.Five diagnostic texture features were included,including three first-order texture features: uniformity,interquartile range,90 th percentile,and two second-order texture feature-gray symbiosis matrix: joint entropy,deficit,with AUC values of0.920,0.920,0.900,0.920 and 0.900,respectively.The final model was constructed from the three highly effective texture features（uniformity,interquartile range,joint entropy）with an AUC value of 0.980.ConclusionsIVIM related parameters（D,f）and texture analysis（texture characteristics: uniformity,interquartile range,90 th percentile,joint entropy,deficit）have differential value for PCa and BPH,which can guide the early clinical intervention of the disease process and improve the patient survival rate and quality of life.The diagnostic value of bp-MRI PI-RADS v2.1 score,age and t PSA for clinically significant prostate cancerObjectiveTo explore the value of bp-MRI PI-RADS v2.1 score,age and t PSA for clinically significant prostate cancer.MethodsPatients who underwent prostate MRI from August 2020 to December 2021 were collected and had definite pathology were collected.In the final study,47 patients were included,including 20 in clinically significant prostate cancer（GS ≥ 7）and 27 in non-clinically significant prostate cancer（GS <7 or prostatic hyperplasia）.The parameters included in the study included age,t PSA values and MRI PI-RADS v2.1score.The parameter values of the two groups.The independent predictors of clinically significant prostate cancer were screened by binary Logistic regression,and the combined model was constructed by combining the good factors.The corresponding ROC curve was drawn and the area under the curve AUC was calculated to evaluate the prediction efficiency.Results1.In the inter-group analysis,the P values for age,t PSA,and PI-RADS v2.1 scores were 0.039,0.000 and 0.000,respectively.2.In the Binary Logistic regression analysis,the P values for age,t PSA and PI-RADS v2.1 scores were 0.186,0.032 and 0.004,respectively;the regression coefficients were0.104,0.282 and 1.549,respectively,and the OR values were 1.110,1.326 and 4.706,respectively.3.The AUC values for the age,t PSA and PI-RADS v2.1 scores were 0.656,0.854 and0.900,respectively.The model was constructed with combined t PSA and PI-RADS v2.1 scores,with an AUC value of 0.965.ConclusionsThe bp-MRI PI-RADS v2.1 score and t PSA have some predictive value for clinically significant prostate cancer,can reduce excessive invasive biopsy and provide direction for clinical diagnosis and treatment,and the combined application of the two has higher efficacy. |
PDF Full Text Request