Prostate Cancer Delineation in MRI Images Based on Deep Learning: Quantitative Comparison and Promising Perspective

Prostate cancer is the most common malignant male tumor. Magnetic Resonance Imaging (MRI) plays a crucial role in the detection, diagnosis, and treatment of prostate cancer diseases. Computer-aided diagnosis systems can help doctors to analyze MRI images and detect prostate cancer earlier. One of the key stages of prostate cancer CAD systems is the automatic delineation of the prostate. Deep learning has recently demonstrated promising segmentation results with medical images. The purpose of this paper is to compare the state-of-the-art of deep learning-based approaches for prostate delineation in MRI images and discussing their limitations and strengths. Besides, we introduce a promising perspective for prostate tumor classification in MRI images. This perspective includes the use of the best segmentation model to detect the prostate tumors in MRI images. Then, we will employ the segmented images to extract the radiomics features that will be used to discriminate benign or malignant prostate tumors.

The clinical guiding value of a radiomics model for androgen deprivation therapy for prostate cancer

Objective Three models were used to evaluate prostate cancer after androgen deprivation therapy (ADT) and to determine the value of detecting residual lesions after treatment. Methods We retrospectively analysed patients with prostate cancer who received ADT from January 2018 to June 2019. Patients were divided into ADT responder and ADT non-responder groups, and clinical risk factors were determined. Regions of interest were manually contoured on each slice on fat-saturated-T2-weighted imaging, and radiomic features were extracted. Uni- and multivariate logistic regression were used to establish radiomics, clinical and combined models. Results There were 23 ADT non-responders and 20 ADT responders. In the clinical model, total prostate-specific antigen concentration and T stage were independent predictors of efficacy (area under the curve (AUC) = 0.774). The characteristics, MinIntensity and Correlation_ angle135_offset4 indicated an effective clinical model (AUC = 0.807). GLCMEntropy_ AllDirection_offset1_SD was the best feature to differentiate residual lesions from the central gland (CG) (Lesion-CG model, AUC = 0.955). Correlation_angle135_offset4, GLCMEntropy_ AllDirection_offset4_SD and GLCMEntropy_AllDirection_offset7_SD differentiated residual lesions from the peripheral zone (PZ) (Lesion-PZ model, AUC = 0.855). The AUC for the combined model was 0.904. Conclusions Our models can guide the clinical treatment of patients with different ADT responses. Furthermore, the radiomics model can detect prostate cancer that is non-responsive to ADT.

A Choosing Wisely top-5 list to support general practitioners in Austria

SummaryFrom a pool of 147 reliable recommendations, ten experts from the Austrian Society of General Practice and Family Medicine selected 21 relevant recommendations as the basis for the Delphi process. In two Delphi rounds, eleven experts established a top‑5 list of recommendations designed for Austrian family practice to reduce medical overuse. Three of the chosen recommendations address the issue of antibiotic usage in patients with viral upper respiratory tract infections, in children with mild otitis media, and in patients with asymptomatic bacteriuria. The other two “do not do” recommendations concern imaging studies for nonspecific low back pain and routine screening to detect prostate cancer. A subsequent survey identified the reasons for selecting these top‑5 recommendations: the frequency of the issue, potential harms, costs, and patients’ expectations. Experts hope the campaign will save time in educating patients and provide legal protection for omitting measures.

Radiomic and Genomic Machine Learning Method Performance for Prostate Cancer Diagnosis: Systematic Literature Review

Background Machine learning algorithms have been drawing attention at the joining of pathology and radiology in prostate cancer research. However, due to their algorithmic learning complexity and the variability of their architecture, there is an ongoing need to analyze their performance. Objective This study assesses the source of heterogeneity and the performance of machine learning applied to radiomic, genomic, and clinical biomarkers for the diagnosis of prostate cancer. One research focus of this study was on clearly identifying problems and issues related to the implementation of machine learning in clinical studies. Methods Following the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) protocol, 816 titles were identified from the PubMed, Scopus, and OvidSP databases. Studies that used machine learning to detect prostate cancer and provided performance measures were included in our analysis. The quality of the eligible studies was assessed using the QUADAS-2 (quality assessment of diagnostic accuracy studies–version 2) tool. The hierarchical multivariate model was applied to the pooled data in a meta-analysis. To investigate the heterogeneity among studies, I2 statistics were performed along with visual evaluation of coupled forest plots. Due to the internal heterogeneity among machine learning algorithms, subgroup analysis was carried out to investigate the diagnostic capability of machine learning systems in clinical practice. Results In the final analysis, 37 studies were included, of which 29 entered the meta-analysis pooling. The analysis of machine learning methods to detect prostate cancer reveals the limited usage of the methods and the lack of standards that hinder the implementation of machine learning in clinical applications. Conclusions The performance of machine learning for diagnosis of prostate cancer was considered satisfactory for several studies investigating the multiparametric magnetic resonance imaging and urine biomarkers; however, given the limitations indicated in our study, further studies are warranted to extend the potential use of machine learning to clinical settings. Recommendations on the use of machine learning techniques were also provided to help researchers to design robust studies to facilitate evidence generation from the use of radiomic and genomic biomarkers.

Attitudes and perceptions towards multiparametric magnetic resonance imaging of the prostate: A national survey among Italian urologists

Objective: We aimed to assess the attitudes and perceptions towards multiparametric magnetic resonance imaging (mpMRI) of the prostate among Italian urologists. Material and Methods: A national, web-based survey was performed. A questionnaire composed of 18 multiple choice questions was e-mailed to 941 currently active urologists, members of the Italian Society of Urology. Preserving anonymity, respondents’ demographics were collected (e.g. geographic region, type of workplace, prostate procedures performed) as well as data concerning their attitudes and perceptions towards mpMRI (e.g. indications deemed appropriate, degree of confidence in mpMRI results). Data were expressed as raw numbers and percentages of survey answers. Results: In total, 98 responses were received (participation rate = 10.4%). Respondents mostly worked in urban areas (96%) and primarily in hospital settings (89%), while 48% of them worked in southern Italy. 97% of respondents considered mpMRI useful to detect Prostate Cancer (PCa) in patients with prior negative biopsy, 64% in biopsy-naïve patients and 60% for PCa pre-operatory staging. About half (42%) of the participants declared that mpMRI results frequently lead them to change PCa management strategy. Standardization of mpMRI acquisition and reporting was partially unsatisfactory. Reported waiting time for mpMRI scans was longer than 4 weeks for 51% of respondents. The major limitation of this survey includes the small number of participants. Conclusions: Prostate mpMRI is used by Italian urologists mainly for detection and for pre-operative staging of PCa. Further improvements in terms of mpMRI availability and report standardization are required.