Magnetic resonance imaging characteristics of chronic prostatitis in patients under the age of 50: is it more than the eye can see?

BACKGROUND: The magnetic resonance (MRI) diagnosis of chronic prostatitis (CP) is insufficiently evaluated. PURPOSE: To evaluate the MRI appearance of CP in young patients by comparing it to individuals with non-prostatic related pathology. MATERIAL AND METHODS: The study included 47 patients with prostatitis-like symptoms evaluated by urologists and referred to pelvic MRI examination (mean age=40.23±7 years; age range=23-49 years) and 93 age-matched individuals with non-prostatic related pathology (mean age=37.5±7 years; age range=21-49 years). All MRI examinations were performed on a 1.5-T machine using a prostate-specific protocol for the prostatitis group and different protocols that included high-resolution small field of view T2-weighted (T2WI) and diffusion-weighted imaging (DWI), for the control group, depending on the clinical indication. RESULTS: Four different T2WI intensity patterns were observed: hyperintense homogenous; slightly to moderate homogenous hypointense; inhomogeneous; and marked hypointense. We found statistically significant differences between the two analyzed groups regarding mean ADC values (P<0.001), distribution of T2WI intensity patterns (P<0.0001), and the presence of dilated venous plexus (P=0.0007). No differences were found regarding prostate volume (P=0.15). In multivariate analysis, all four analyzed imaging parameters were independent predictors of chronic prostatitis (R2=0.67; P<0.0001). Considered together, an age >28 years, an inhomogeneous or marked hypointense T2WI intensity pattern (types 3 and 4), an ADC value ≤1250, and the presence of dilated venous plexus are able to predict CP with an AUC of 93% (sensitivity=85.1%, specificity=88.4%). CONCLUSION: MR parameters like T2WI intensity patterns, ADC values, and venous plexus appearance are promising non-invasive tools in the challenging environment of CP diagnosis.

Fully automated bladder tumor segmentation from T2 MRI images using 3D U-Net algorithm.

Introduction: Bladder magnetic resonance imaging (MRI) has been recently integrated in the diagnosis pathway of bladder cancer. However, automatic recognition of suspicious lesions is still challenging. Thus, development of a solution for proper delimitation of the tumor and its separation from the healthy tissue is of primordial importance. As a solution to this unmet medical need, we aimed to develop an artificial intelligence-based decision support system, which automatically segments the bladder wall and the tumor as well as any suspect area from the 3D MRI images. Materials: We retrospectively assessed all patients diagnosed with bladder cancer, who underwent MRI at our department (n=33). All examinations were performed using a 1.5 Tesla MRI scanner. All images were reviewed by two radiologists, who performed manual segmentation of the bladder wall and all lesions. First, the performance of our fully automated end-to-end segmentation model based on a 3D U-Net architecture (by considering various depths of 4, 5 or 6 blocks) trained in two data augmentation scenarios (on 5 and 10 augmentation datasets per original data, respectively) was tested. Second, two learning setups were analyzed by training the segmentation algorithm with 7 and 14 MRI original volumes, respectively. Results: We obtained a Dice-based performance over 0.878 for automatic segmentation of bladder wall and tumors, as compared to manual segmentation. A larger training dataset using 10 augmentations for 7 patients could further improve the results of the U-Net-5 model (0.902 Dice coefficient at image level). This model performed best in terms of automated segmentation of bladder, as compared to U-Net-4 and U-Net-6. However, in this case increased time for learning was needed as compared to U-Net-4. We observed that an extended dataset for training led to significantly improved segmentation of the bladder wall, but not of the tumor. Conclusion: We developed an intelligent system for bladder tumors automated diagnostic, that uses a deep learning model to segment both the bladder wall and the tumor. As a conclusion, low complexity networks, with less than five-layers U-Net architecture are feasible and show good performance for automatic 3D MRI image segmentation in patients with bladder tumors.

Long-term oncologic and functional outcomes after robot-assisted partial nephrectomy in elderly patients.

BACKGROUND: The aim of this study was to assess the long-term oncologic and functional outcomes in elderly patients having undergone robot-assisted partial nephrectomy (RAPN) for renal cancer (RC). METHODS: Sixty-one patients out of 323 who underwent RAPN for localized RC between July 2009 and March 2016 in our high-volume robotic surgery center (>800 procedures/year), had 70 years or more. Inclusion criteria of the study were age ≥70 years; pathological confirmed RCC and ASA Score ≤3. All patients were stratified according to PADUA classification system in three groups: <7 points, 8-9 points, >10 points. Trifecta was defined as a warm ischemia time (WIT) less then 25 min, negative surgical margins and no perioperative complications. RESULTS: A total of 52 patients were included; median follow-up was 47 months. Median age was 74 yrs. (IQR 72-76.5). Complication rate was 15.4%. Trifecta failure was associated to PADUA Score (P=0.02), and tumor diameter (P=0.04). Renal function was altered in 10 (19.2%) patients before surgery and at last follow-up in 11 (21.1%) patients (CKD stage>2) The DFS, OS and CSS were 89.33%, 90.06% and 94.4%, respectively. CONCLUSIONS: In a high-volume center, robot-assisted approach is feasible and safe in surgical fit elderly patients with good long-term oncologic outcomes.

Investigation of Renal Cell Carcinoma by Contrast-Enhanced Ultrasound- Predictive Value of Time Intensity Curve Analysis in Establishing Local Tumor Invasion and Stage: A Pilot Study.

PURPOSE: Contrast-enhanced ultrasound (CEUS) allows for real-time examination of signal intensity changes in a region of interest (ROI) and quantification of contrast agent kinetics. This study assessed the predictive ability of time-intensity curve (TIC) parameters for local tumor invasion and T stage of renal cell carcinoma (RCC). MATERIALS AND METHODS: Renal tumors in 41 patients were examined by CEUS. Thirty-two met the inclusion criteria, with a total of 33 tumors (27 clear cell, 4 chromophobe, and 2 papillary type I). Nineteen (57.6%) tumors were included in group A (stages pT1 and pT2) and 14 (42.4%) in group B (stage pT3). ROIs were established as: whole tumor (TuW); tumor area with the highest signal intensity (TuMAX) and renal cortex (Ref). The TIC param­eters for each ROI were calculated as below: peak signal intensity, time to peak (TTP), rise time (RT), and mean transit time (MTT). They were analyzed as a whole value for each ROI and as a ratio between the different ROIs. RESULTS: There were significant differences between the tumors invading and not invading the renal sinus fat for TTP (TuW/Ref) [0.98 (0.67-1.25) vs. 1.18 (1.08-1.3), P < .05]. For differentiation between groups A and B, the following ratios were proven as predictors by univariate regression analysis: TTP (TuMAX/TuW); MTT (Tu­MAX/TuW); RT (TuMAX/TuW) (P = .03, P = .01 and P = .02, respectively). The value derived from the Receiver Operating Characteristic (ROC) curve for RT (TuMAX/TuW) was 0.8 with sensitivity = 78.6%, specificity = 89.5%, and cutoff value of > 0.91. CONCLUSION: TIC parameters were predictors of locally noninvasive and invasive RCC.