Multi-parametric MRI (mp-MRI) is considered the best non-invasive imaging modality for diagnosing prostate cancer (PCa). However, mp-MRI for PCa diagnosis is currently limited by the qualitative or semi-quantitative interpretation criteria, leading to inter-reader variability and a suboptimal ability to assess lesion aggressiveness. Convolutional neural networks (CNNs) are a powerful method to automatically learn the discriminative features for various tasks, including cancer detection. We propose a novel multi-class CNN, FocalNet, to jointly detect PCa lesions and predict their aggressiveness using Gleason score (GS). FocalNet characterizes lesion aggressiveness and fully utilizes distinctive knowledge from mp-MRI. We collected a prostate mp-MRI dataset from 417 patients who underwent 3T mp-MRI exams prior to robotic-assisted laparoscopic prostatectomy (RALP). FocalNet is trained and evaluated in this large study cohort with 5-fold cross-validation. In the free-response receiver operating characteristics (FROC) analysis for lesion detection, FocalNet achieved 89.7% and 87.9% sensitivity for index lesions and clinically significant lesions at 1 false positive per patient, respectively. For GS classification, evaluated by the receiver operating characteristics (ROC) analysis, FocalNet received the area under the curve (AUC) of 0.81 and 0.79 for the classifications of clinically significant PCa (GS≥3+4) and PCa with GS≥4+3, respectively. With the comparison to the prospective performance of radiologists using the current diagnostic guideline, FocalNet demonstrated comparable detection sensitivity for index lesions and clinically significant lesions, only 3.4% and 1.5% lower than highly experienced radiologists without statistical significance.

No abstract

GENITOURINARY IMAGINGM ultiparametric MRI is an important tool in the diagnosis of prostate cancer (PCa) (1,2). However, multiparametric MRI still misses PCa in up to 45% of men and faces challenges in distinguishing clinically significant PCa from indolent PCa (2,3). Thus, histopathologic examination of PCa remains the reference standard. A Gleason score based on the microscopic appearance of PCa is assigned to indicate its aggressiveness (4).Diffusion-weighted MRI is a critical component of multiparametric MRI and is sensitive to tissue microstructure changes in PCa (5). However, current clinical analysis using a monoexponential signal model to calculate apparent diffu-Materials and Methods: Men with PCa who underwent 3-T MRI and robotic-assisted radical prostatectomy between June 2018 and January 2019 were prospectively studied. After prostatectomy, the fresh whole prostate specimens were imaged in patient-specific threedimensionally printed molds by using 3-T MRI with DR-CSI and were then sliced to create coregistered WMHP slides. The DR-CSI spectral signal component fractions (f A , f B , f C ) were compared with epithelial, stromal, and luminal area fractions (f epithelium , f stroma , f lumen ) quantified in PCa and benign tissue regions. A linear mixed-effects model assessed the correlations between (f A , f B , f C ) and (f epithelium , f stroma , f lumen ), and the strength of correlations was evaluated by using Spearman correlation coefficients. Differences between PCa and benign tissues in terms of DR-CSI signal components and microscopic tissue compartments were assessed using two-sided t tests.Results: Prostate specimens from nine men (mean age, 65 years 6 7 [standard deviation]) were evaluated; 20 regions from 17 PCas, along with 20 benign tissue regions of interest, were analyzed. Three DR-CSI spectral signal components (spectral peaks) were consistently identified. The f A , f B , and f C were correlated with f epithelium , f stroma , and f lumen (all P , .001), with Spearman correlation coefficients of 0.74 (95% confidence interval [CI]: 0.62, 0.83), 0.80 (95% CI: 0.66, 0.89), and 0.67 (95% CI: 0.51, 0.81), respectively. PCa exhibited differences compared with benign tissues in terms of increased f A (PCa vs benign, 0.37 6 0.05 vs 0.27 6 0.06; P , .001), decreased f C (PCa vs benign, 0.18 6 0.06 vs 0.31 6 0.13; P = .01), increased f epithelium (PCa vs benign, 0.44 6 0.13 vs 0.26 6 0.16; P , .001), and decreased f lumen (PCa vs benign, 0.14 6 0.08 vs 0.27 6 0.18; P = .004). Conclusion:Diffusion-relaxation correlation spectrum imaging signal components correlate with microscopic tissue compartments in the prostate and differ between cancer and benign tissue.

No abstract

BACKGROUND: Hemiablation is a less morbid treatment alternative for appropriately selected patients with unilateral prostate cancer (PCa). However, to the authors' knowledge, traditional diagnostic techniques inadequately identify appropriate candidates. In the current study, the authors quantified the accuracy for identifying hemiablation candidates using contemporary diagnostic techniques, including multiparametric magnetic resonance imaging (mpMRI) and MRI-fusion with complete systematic template biopsy. METHODS: A retrospective analysis of patients undergoing MRI and MRI-fusion prostate biopsy, including full systematic template biopsy, prior to radical prostatectomy in a single tertiary academic institution between June 2010 and February 2018 was performed. Hemiablation candidates had unilateral intermediate-risk PCa (Gleason score [GS] of 3+4 or 4+3, clinical T classification ≤T2, and prostate-specific antigen level <20 ng/dL) on MRI-fusion biopsy and 2) no contralateral highly or very highly suspicious Prostate Imaging Reporting and Data System version 2 (PI-RADSv2) MRI lesions. Hemiablation candidates were inappropriately selected if pathologists identified contralateral GS ≥3+4 or high-risk ipsilateral PCa on prostatectomy. The authors tested a range of hemiablation inclusion criteria and performed multivariable analysis of preoperative predictors of undetected contralateral disease. RESULTS: Of 665 patients, 92 met primary hemiablation criteria. Of these 92 patients, 44 (48%) were incorrectly identified due to ipsilateral GS ≥3+4 tumors crossing the midline (21 patients), undetected distinct contralateral GS ≥3+4 tumors (20 patients), and/ or ipsilateral high-risk PCa (3 patients) on prostatectomy. The rate of undetected contralateral disease ranged from 41% to 48% depending on inclusion criteria. On multivariable analysis, men with anterior index tumors were found to be 2.4 times more likely to harbor undetected contralateral GS ≥3+4 PCa compared with men with posterior lesions (P < .05). CONCLUSIONS: Clinicians and patients must weigh the risk of inadequate oncologic treatment against the functional benefits of hemiablation. Further investigation into methods for improving patient selection for hemiablation is necessary. Cancer 2019;125:2955-2964.