Visual inspection of histopathology images of stained biopsy tissue by expert pathologists is the standard method for grading of prostate cancer (PCa). However, this process is time-consuming and subject to high inter-observer variability. Machine learning-based methods have the potential to improve efficient throughput of large volumes of slides while decreasing variability, but they are not easy to develop because they require substantial amounts of labeled training data. In this paper, we propose a deep learning-based classification technique and data augmentation methods for accurate grading of PCa in histopathology images in the presence of limited data. Our method combines the predictions of three separate convolutional neural networks (CNNs) that work with different patch sizes. This enables our method to take advantage of the greater amount of contextual information in larger patches as well as greater quantity of smaller patches in the labeled training data. The predictions produced by the three CNNs are combined using a logistic regression model, which is trained separately after the CNN training. To effectively train our models, we propose new data augmentation methods and empirically study their effects on the classification accuracy. The proposed method achieves an accuracy of 92 % in classifying cancerous patches versus benign patches and an accuracy of 90 % in classifying low-grade (i.e., Gleason grade 3) from high-grade (i.e., Gleason grades 4 and 5) patches. The agreement level of our automatic grading method with expert pathologists is within the range of agreement between pathologists. Our experiments indicate that data augmentation is necessary for achieving expert-level performance with deep learning-based methods. A combination of image-space augmentation and feature-space augmentation leads to the best results. Our study shows that well-designed and properly trained deep learning models can achieve PCa Gleason grading accuracy that is comparable to an expert pathologist.

Introduction: There is limited, yet compelling evidence supporting the role of surgeon technical performance in influencing patient outcomes. To date, this concept has been underexplored in endourologic procedures. We hypothesized that a surgeon's technical performance plays a role in predicting an early return to continence after robot-assisted radical prostatectomy (RARP). Materials and Methods: We conducted a retrospective, matched case–control analysis of prospectively collected unedited RARP endoscopic videos performed by a single surgeon. A blinded observer with expertise in intraoperative video analysis evaluated clinically relevant steps of RARP using the global evaluative assessment of robotic skill (GEARS) and the generic error rating tool (GERT). The primary outcome was continence status at 3 months postoperatively, defined as patient use of less than or equal to a single precautionary pad. Mann–Whitney U tests examined differences in predictor variables between cases and controls, and multivariate analysis was conducted using binary logistic regression models. Results: Twenty-four incontinent patients were matched for age, body mass index, preoperative International Prostate Symptoms Score, use of posterior/anterior hitch, prostate weight, and learning curve position. No statistically significant difference in errors between groups was observed using the GERT. On multivariate analysis, overall case GEARS score was independently predictive of 3-month continence status (odds ratios [OR] = 0.55, 95% confidence interval [CI] 0.33–0.91), as were urethrovesical anastomosis (OR = 0.70, 95% CI 0.50–0.97) and bladder neck GEARS scores (OR = 0.69, 95% CI 0.51–0.94). Conclusions: Our study generates the hypothesis that there may be a link between surgeon technical performance and functional outcomes in RARP. This relationship may have implications for the accreditation and training of future urologists and warrants further investigation.

The number of prostate cancer survivors is rapidly growing in the Western world. As a result of better oncologic outcomes, more patients are living longer with the adverse effects of treatment, which can be both functional and psychological. Clinicians, in an era of shared decision-making, must not only cure the cancer, but also ensure that, after treatment, their patients experience the best quality of life and minimal post-treatment decisional regret. To participate in the decision-making process, men and their involved partners and family need to fully understand the relative benefits and harms of prostate cancer treatments. Patient preference studies indicate that men with prostate cancer are not well informed. Decision- making aids are a positive treatment adjunct both to convey information and to allow patients to explore their own beliefs and values during the decision-making process. The evidence suggests that decision-making aids better prepare patients for involvement in treatment decisions, but further studies are required to investigate the relationship between the use of decision-making aids and post-treatment decisional regret in prostate cancer.

Graph convolutional neural networks have shown significant potential in natural and histopathology images. However, their use has only been studied in a single magnification or multi-magnification with either homogeneous graphs or only different node types. In order to leverage the multi-magnification information and improve message passing with graph convolutional networks, we handle different embedding spaces at each magnification by introducing the Multi-Scale Relational Graph Convolutional Network (MS-RGCN) as a multiple instance learning method. We model histopathology image patches and their relation with neighboring patches and patches at other scales (i.e., magnifications) as a graph. We define separate message-passing neural networks based on node and edge types to pass the information between different magnification embedding spaces. We experiment on prostate cancer histopathology images to predict the grade groups based on the extracted features from patches. We also compare our MS-RGCN with multiple state-of-the-art methods with evaluations on several source and held-out datasets. Our method outperforms the state-of-the-art on all of the datasets and image types consisting of tissue microarrays, whole-mount slide regions, and whole-slide images. Through an ablation study, we test and show the value of the pertinent design features of the MS-RGCN.