Project description:ObjectiveTo assess the outcomes of pre-biopsy magnetic resonance imaging (MRI) pathways, as a tool in biopsy-naïve men with suspicion of prostate cancer, in routine clinical practice. Secondary outcomes included a comparison of transrectal MRI-directed biopsy (TR-MRDB) and transperineal (TP)-MRDB in men with suspicious MRI.Patients and methodsWe retrospectively assessed a two-centre cohort of consecutive biopsy-naïve men with suspicion of prostate cancer who underwent a Prostate Imaging-Reporting and Data System version 2 (PI-RADS v2) compliant pre-biopsy MRI in a single, high-volume centre between 2015 and 2019 (Centre 1). Men with suspicious MRI scans underwent TR-MRDB in Centre 1 and TP-MRDB with additional random biopsies (RB) in Centre 2. The MRI and histopathology were assessed in the same institution (Centre 1). Outcomes included: (i) overall detection rates of Grade Group (GG) 1, GG ≥2, and GG ≥3 cancer in men with suspicious MRI; (ii) Biopsy-avoidance due to non-suspicious MRI; and (iii) Cancer detection rates and biopsy-related complications between TR- and TP-MRDB. To reduce confounding bias for MRDB comparisons, inverse probability weighting (IPW) was performed for age, digital rectal examination, prostate-specific antigen (PSA), prostate volume, PSA density, and PI-RADS category.ResultsOf the 2597 men included, the overall GG 1, GG ≥2, and GG ≥3 prevalence was 8% (210/2597), 27% (697/2597), and 15% (396/2597), respectively. Biopsy was avoided in 57% (1488/2597) of men. After IPW, the GG 1, GG ≥2 and GG ≥3 detection rates after TR- and TP-MRDB were comparable at 24%, 57%, and 32%; and 18%, 64%, and 38%, respectively; with mean differences of -5.7% (95% confidence interval [CI] -13% to 1.4%), 6.1% (95% CI -2.1% to 14%), and 5.7% (95% CI -1.7% to 13%). Complications were similar in TR-MRDB (0.50%) and TP-MRDB with RB (0.62%; mean difference 0.11%, 95% CI -0.87% to 1.1%).ConclusionThis high-volume, two-centre study shows pre-biopsy MRI as a decision tool is implementable in daily clinical practice. Compared to recent trials, a substantially higher biopsy avoidance rate was achieved without compromising GG ≥2/GG ≥3 detection and coinciding with lower over detection rates of GG 1 cancer. Prostate cancer detection and complication rates were comparable for TR- and TP-MRDB.

Project description:BackgroundSynthetic magnetic resonance imaging (SyMRI) is a fast, standardized, and robust novel quantitative technique that has the potential to circumvent the subjectivity of interpretation in prostate multiparametric magnetic resonance imaging (mpMRI) and the limitations of existing MRI quantification techniques. Our study aimed to evaluate the potential utility of SyMRI in the diagnosis and aggressiveness assessment of prostate cancer (PCA).MethodsWe retrospectively analyzed 309 patients with suspected PCA who had undergone mpMRI and SyMRI, and pathologic results were obtained by biopsy or PCA radical prostatectomy (RP). Pathological types were classified as PCA, benign prostatic hyperplasia (BPH), or peripheral zone (PZ) inflammation. According to the Gleason Score (GS), PCA was divided into groups of intermediate-to-high risk (GS ≥4+3) and low-risk (GS ≤3+4). Patients with biopsy-confirmed low-risk PCA were further divided into upgraded and nonupgraded groups based on the GS changes of the RP results. The values of the apparent diffusion coefficient (ADC), T1, T2 and proton density (PD) of these lesions were measured on ADC and SyMRI parameter maps by two physicians; these values were compared between PCA and BPH or inflammation, between the intermediate-to-high-risk and low-risk PCA groups, and between the upgraded and nonupgraded PCA groups. The risk factors affecting GS grades were identified via univariate analysis. The effects of confounding factors were excluded through multivariate logistic regression analysis, and independent predictive factors were calculated. Subsequently, the ADC+Sy(T2+PD) combined models for predicting PCA risk grade or GS upgrade were constructed through data processing analysis. The diagnostic performance of each parameter and the ADC+Sy(T2+PD) model was analyzed. The calibration curve was calculated by the bootstrapping internal validation method (200 bootstrap resamples).ResultsThe T1, T2, and PD values of PCA were significantly lower than those of BPH or inflammation (P≤0.001) in both the PZ or transitional zone. Among the 178 patients with PCA, intermediate-to-high-risk PCA group had significantly higher T1, T2, and PD values but lower ADC values compared with the low-risk group (P<0.05), and the diagnostic efficacy of each single parameter was similar (P>0.05). The ADC+Sy(T2+PD) model showed the best performance, with an area under the curve (AUC) 0.110 [AUC =0.818; 95% confidence interval (CI): 0.754-0.872] higher than that of ADC alone (AUC =0.708; 95% CI: 0.635-0.774) (P=0.003). Among the 68 patients initially classified as PCA in the low-risk group by biopsy, PCA in the postoperative upgraded GS group had significantly higher T1, T2, and PD values but lower ADC values than did those in the nonupgraded group (P<0.01). In addition, the ADC+Sy(T2+PD) model better predicted the upgrade of GS, with a significant increase in AUC of 0.204 (AUC =0.947; 95% CI: 0.864-0.987) compared with ADC alone (AUC =0.743; 95% CI: 0.622-0.841) (P<0.001).ConclusionsQuantitative parameters (T1, T2, and PD) derived from SyMRI can help differentiate PCA from non-PCA. Combining SyMRI parameters with ADC significantly improved the ability to differentiate between intermediate-to-high risk PCA from low-risk PCA and could predict the upgrade of low-risk PCA as confirmed by biopsy.

Project description:Background: Multiparametric magnetic resonance imaging (mpMRI) plays an important role in the diagnosis of prostate cancer (PCa) in the current clinical setting. However, the performance of mpMRI usually varies based on the experience of the radiologists at different levels; thus, the demand for MRI interpretation warrants further analysis. In this study, we developed a deep learning (DL) model to improve PCa diagnostic ability using mpMRI and whole-mount histopathology data. Methods: A total of 739 patients, including 466 with PCa and 273 without PCa, were enrolled from January 2017 to December 2019. The mpMRI (T2 weighted imaging, diffusion weighted imaging, and apparent diffusion coefficient sequences) data were randomly divided into training (n = 659) and validation datasets (n = 80). According to the whole-mount histopathology, a DL model, including independent segmentation and classification networks, was developed to extract the gland and PCa area for PCa diagnosis. The area under the curve (AUC) were used to evaluate the performance of the prostate classification networks. The proposed DL model was subsequently used in clinical practice (independent test dataset; n = 200), and the PCa detective/diagnostic performance between the DL model and different level radiologists was evaluated based on the sensitivity, specificity, precision, and accuracy. Results: The AUC of the prostate classification network was 0.871 in the validation dataset, and it reached 0.797 using the DL model in the test dataset. Furthermore, the sensitivity, specificity, precision, and accuracy of the DL model for diagnosing PCa in the test dataset were 0.710, 0.690, 0.696, and 0.700, respectively. For the junior radiologist without and with DL model assistance, these values were 0.590, 0.700, 0.663, and 0.645 versus 0.790, 0.720, 0.738, and 0.755, respectively. For the senior radiologist, the values were 0.690, 0.770, 0.750, and 0.730 vs. 0.810, 0.840, 0.835, and 0.825, respectively. The diagnosis made with DL model assistance for radiologists were significantly higher than those without assistance (P < 0.05). Conclusion: The diagnostic performance of DL model is higher than that of junior radiologists and can improve PCa diagnostic accuracy in both junior and senior radiologists.

Project description:ObjectiveProstate lesions detected with multiparametric magnetic resonance imaging (mpMRI) are classified for their malignant potential according to the Prostate Imaging-Reporting And Data System (PI-RADS™2). In this study, we evaluate the diagnostic accuracy of the mpMRI with and without gadolinium, with emphasis on the added diagnostic value of the dynamic contrast enhancement (DCE).Materials and methodsThe study was retrospective for 286 prostate lesions / 213 eligible patients, n = 116/170, and 49/59% malignant for the peripheral (Pz) and transitional zone (Tz), respectively. A stereotactic MRI-guided prostate biopsy served as the histological ground truth. All patients received a mpMRI with DCE. The influence of DCE in the prediction of malignancy was analyzed by blinded assessment of the imaging protocol without DCE and the DCE separately.ResultsSignificant (CSPca) and insignificant (IPca) prostate cancers were evaluated separately to enhance the potential effects of the DCE in the detection of CSPca. The Receiver Operating Characteristics Area Under Curve (ROC-AUC), sensitivity (Se) and specificity (Spe) of PIRADS-without-DCE in the Pz was 0.70/0.47/0.86 for all cancers (IPca and CSPca merged) and 0.73/0.54/0.82 for CSPca. PIRADS-with-DCE for the same patients showed ROC-AUC/Se/Spe of 0.70/0.49/0.86 for all Pz cancers and 0.69/0.54/0.81 for CSPca in the Pz, respectively, p>0.05 chi-squared test. Similar results for the Tz, AUC/Se/Spe for PIRADS-without-DCE was 0.75/0.61/0.79 all cancers and 0.67/0.54/0.71 for CSPca, not influenced by DCE (0.66/0.47/0.81 for all Tz cancers and 0.61/0.39/0.75 for CSPca in Tz). The added Se and Spe of DCE for the detection of CSPca was 88/34% and 78/33% in the Pz and Tz, respectively.ConclusionDCE showed no significant added diagnostic value and lower specificity for the prediction of CSPca compared to the non-enhanced sequences. Our results support that gadolinium might be omitted without mitigating the diagnostic accuracy of the mpMRI for prostate cancer.

Project description:ObjectiveTo perform a meta-analysis to quantitatively assess functional magnetic resonance imaging (MRI) in the diagnosis of locally recurrent prostate cancer.Materials and methodsA comprehensive search of the PubMed, Embase, Cochrane Central Register of Controlled Trials, and Cochrane Database of Systematic Reviews was conducted from January 1, 1995 to December 31, 2016. Diagnostic accuracy was quantitatively pooled for all studies by using hierarchical logistic regression modeling, including bivariate modeling and hierarchical summary receiver operating characteristic (HSROC) curves (AUCs). The Z test was used to determine whether adding functional MRI to T2-weighted imaging (T2WI) results in significantly increased diagnostic sensitivity and specificity.ResultsMeta-analysis of 13 studies involving 826 patients who underwent radical prostatectomy showed a pooled sensitivity and specificity of 91%, and the AUC was 0.96. Meta-analysis of 7 studies involving 329 patients who underwent radiotherapy showed a pooled sensitivity of 80% and specificity of 81%, and the AUC was 0.88. Meta-analysis of 11 studies reporting 1669 sextant biopsies from patients who underwent radiotherapy showed a pooled sensitivity of 54% and specificity of 91%, and the AUC was 0.85. Sensitivity after radiotherapy was significantly higher when diffusion-weighted MRI data were combined with T2WI than when only T2WI results were used. This was true when meta-analysis was performed on a per-patient basis (p = 0.027) or per sextant biopsy (p = 0.046). A similar result was found when 1H-magnetic resonance spectroscopy (1H-MRS) data were combined with T2WI and sextant biopsy was the unit of analysis (p = 0.036).ConclusionFunctional MRI data may not strengthen the ability of T2WI to detect locally recurrent prostate cancer in patients who have undergone radical prostatectomy. By contrast, diffusion-weight MRI and 1H-MRS data may improve the sensitivity of T2WI for patients who have undergone radiotherapy.

Project description:Objectives: To evaluate a new deep neural network (DNN)-based computer-aided diagnosis (CAD) method, namely, a prostate cancer localization network and an integrated multi-modal classification network, to automatically localize prostate cancer on multi-parametric magnetic resonance imaging (mp-MRI) and classify prostate cancer and non-cancerous tissues. Materials and methods: The PROSTAREx database consists of a "training set" (330 suspected lesions from 204 cases) and a "test set" (208 suspected lesions from 104 cases). Sequences include T2-weighted, diffusion-weighted, Ktrans, and apparent diffusion coefficient (ADC) images. For the task of abnormal localization, inspired by V-net, we designed a prostate cancer localization network with mp-MRI data as input to achieve automatic localization of prostate cancer. Combining the concepts of multi-modal learning and ensemble learning, the integrated multi-modal classification network is based on the combination of mp-MRI data as input to distinguish prostate cancer from non-cancerous tissues through a series of operations such as convolution and pooling. The performance of each network in predicting prostate cancer was examined using the receiver operating curve (ROC), and the area under the ROC curve (AUC), sensitivity (TPR), specificity (TNR), accuracy, and Dice similarity coefficient (DSC) were calculated. Results: The prostate cancer localization network exhibited excellent performance in localizing prostate cancer, with an average error of only 1.64 mm compared to the labeled results, an error of about 6%. On the test dataset, the network had a sensitivity of 0.92, specificity of 0.90, PPV of 0.91, NPV of 0.93, and DSC of 0.84. Compared with multi-modal classification networks, the performance of single-modal classification networks is slightly inadequate. The integrated multi-modal classification network performed best in classifying prostate cancer and non-cancerous tissues with a TPR of 0.95, TNR of 0.82, F1-Score of 0.8920, AUC of 0.912, and accuracy of 0.885, which fully confirmed the feasibility of the ensemble learning approach. Conclusion: The proposed DNN-based prostate cancer localization network and integrated multi-modal classification network yielded high performance in experiments, demonstrating that the prostate cancer localization network and integrated multi-modal classification network can be used for computer-aided diagnosis (CAD) of prostate cancer localization and classification.

Dataset Information

Editorial: Magnetic resonance imaging applications for the diagnosis of prostate cancer

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets