3 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2018.

Background: To evaluate the potential of clinical-based model, a biparametric MRI-based radiomics model and a clinical-radiomics combined model for predicting clinically significant prostate cancer (PCa).Methods: In total, 381 patients with clinically suspicious PCa were included in this retrospective study; of those, 199 patients did not have PCa upon biopsy, while 182 patients had PCa. All patients underwent 3.0-T MRI examinations with the same acquisition parameters, and clinical risk factors associated with PCa (age, prostate volume, serum PSA, etc.) were collected. We randomly stratified the training and test sets using a 6:4 ratio. The radiomic features included gradient-based histogram features, grey-level co-occurrence matrix (GLCM), run-length matrix (RLM), and grey-level size zone matrix (GLSZM). Three models were developed using multivariate logistic regression analysis to predict clinically significant PCa: a clinical model, a radiomics model and a clinical-radiomics combined model. The diagnostic performance and clinical net benefit of each model were compared via receiver operating characteristic (ROC) curve analysis and decision curves, respectively.Results: Both the radiomics model (AUC: 0.98) and the clinical-radiomics combined model (AUC: 0.98) achieved greater predictive efficacy than the clinical model (AUC: 0.79). The decision curve analysis also showed that the radiomics model and combined model had higher net benefits than the clinical model. Conclusions:Compared with the evaluation of clinical risk factors associated with PCa only, the radiomics-based machine learning model can improve the predictive accuracy for clinically significant PCa, in terms of both diagnostic performance and clinical net benefit.

No abstract

No abstract

Background: Prostate Imaging Reporting and Data System version 2 (PI-RADS V2) 3 category lesions are of intermediate status with an equivocal risk of presenting clinically significant prostate cancer (csPCa).How to avoid excessive biopsies while improving the csPCa detection rate in these lesions has always been a clinical problem that needed to be solved. The purpose of this study is to explore the csPCa diagnostic value of clinical and magnetic resonance imaging (MRI) data for peripheral and transitional zone (PZ and TZ, respectively) PI-RADS 3 lesions to aid in clinical decision-making and reduce excessive biopsies. Methods:From March 2016 to October 2018, a total of 629 men who underwent a prostate MRI and subsequently biopsy were enrolled. Two radiologists (with 3 and 7 years of experience, respectively) independently reviewed and scored all images using the PI-RADS V2 scoring criteria. Clinical and MRI data of men with PI-RADS 3 index lesions were collected by another radiologist. Univariate and multivariate analyses were performed to identify the risk factors of csPCa.Results: In a subset of 121 men with 121 PI-RADS 3 index lesions, 25.6% of the lesions (31/121) were PCa (Gleason score ≥6), and 11.6% (14/121) were csPCa (Gleason score ≥7). Further, 44.6% of lesions (54/121) were located in the PZ and 55.4% (67/121) in the TZ. For PZ lesions, 18.5% of the lesions (10/54) were csPCa. Prostate-specific antigen density (PSAD) (P=0.024) and age (P=0.026) were independent risk factors for csPCa in the multivariate logistic analysis. The combination of PSAD and age yielded an area under the curve (AUC) value of 0.816 for predicting csPCa. If biopsy had been restricted to patients with a PSAD ≥0.15 ng/mL 2 or an age >68 years, 24.1% (13/54) of patients would have avoided biopsy but only 1 (10%) csPCa would have been missed, with a sensitivity of 80.0% and negative predictive value (NPV) of 92.3%. For TZ lesions, only 6.0% of the lesions (4/67) were csPCa. The PSA and PSAD values in the PI-RADS 3 TZ lesions were higher in the csPCa group (45.07 and 0.47 ng/mL 2 , respectively) than in the non-csPCa group (10.03 and 0.17 ng/mL 2 , respectively). Conclusions:CsPCa was detected at a relatively high rate in PI-RADS 3 PZ lesions. Combining PSAD and age could help to reduce excessive biopsies of such lesions. CsPCa is unlikely to be detected in PI-RADS 3 TZ lesions; thus, active surveillance may be an optimal choice for these lesions, especially among patients without high-risk factors.