RESUMEN
Purpose@#To explore the role of artificial intelligence and machine learning (ML) techniques in oncological urology. In recent years, our group investigated the prostate cancer gene 3 (PCA3) score, prostate-specific antigen (PSA), and free-PSA predictive role for prostate cancer (PCa), using the classical binary logistic regression (LR) modeling. In this research, we approached the same clinical problem by several different ML algorithms, to evaluate their performances and feasibility in a real-world evidence PCa detection trial. @*Materials and Methods@#The occurrence of a positive biopsy has been studied in a large cohort of 1,246 Italian men undergoing first or repeat biopsy. Seven supervised ML algorithms were selected to build biomarkers-based predictive models: generalized linear model, gradient boosting machine, eXtreme gradient boosting machine (XGBoost), distributed random forest/ extremely randomized forest, multilayer artificial Deep Neural Network, naïve Bayes classifier, and an automatic ML ensemble function. @*Results@#All the ML models showed better performances in terms of area under curve (AUC) and accuracy, when compared to LR model. Among them, an XGBoost model tuned by the autoML function reached the best metrics (AUC, 0.830), well overtaking LR results (AUC, 0.738). In the variable importance ranking coming from this XGBoost model (accuracy, 0.824), the PCA3 score importance was 3-fold and 4-fold larger, when compared to that of free-PSA and PSA, respectively. @*Conclusions@#The ML approach proved to be feasible and able to achieve good predictive performances with reproducible results: it may thus be recommended, when applied to PCa prediction based on biomarkers fluctuations.
RESUMEN
Purpose@#To explore the role of artificial intelligence and machine learning (ML) techniques in oncological urology. In recent years, our group investigated the prostate cancer gene 3 (PCA3) score, prostate-specific antigen (PSA), and free-PSA predictive role for prostate cancer (PCa), using the classical binary logistic regression (LR) modeling. In this research, we approached the same clinical problem by several different ML algorithms, to evaluate their performances and feasibility in a real-world evidence PCa detection trial. @*Materials and Methods@#The occurrence of a positive biopsy has been studied in a large cohort of 1,246 Italian men undergoing first or repeat biopsy. Seven supervised ML algorithms were selected to build biomarkers-based predictive models: generalized linear model, gradient boosting machine, eXtreme gradient boosting machine (XGBoost), distributed random forest/ extremely randomized forest, multilayer artificial Deep Neural Network, naïve Bayes classifier, and an automatic ML ensemble function. @*Results@#All the ML models showed better performances in terms of area under curve (AUC) and accuracy, when compared to LR model. Among them, an XGBoost model tuned by the autoML function reached the best metrics (AUC, 0.830), well overtaking LR results (AUC, 0.738). In the variable importance ranking coming from this XGBoost model (accuracy, 0.824), the PCA3 score importance was 3-fold and 4-fold larger, when compared to that of free-PSA and PSA, respectively. @*Conclusions@#The ML approach proved to be feasible and able to achieve good predictive performances with reproducible results: it may thus be recommended, when applied to PCa prediction based on biomarkers fluctuations.
RESUMEN
Objective: Extended pelvic lymph nodes dissection (EPLND) allows the removal of a higher number of lymph nodes than limited PLND. The aims of this study were to describe our robot-assisted EPLND (RAEPLND) technique with related complications, and to report the number of lymph nodes removed and the rate of lymph nodal metastasis. Materials and Methods: 153 patients underwent RAEPLND prior to robot-assisted radical prostatectomy (RARP). Indications were defined according to Briganti nomogram, to predict risk of lymph-nodal metastasis. Lymphatic packages covering the distal tract of the common iliac artery, the medial portion of the external iliac artery, the external iliac vein and the internal iliac vessels, together with the obturator and the presacral lymphatic packages were removed on both sides. Results: Median preoperative PSA was 7.5 ng/mL (IQR 5.5–11.5). Median operative time was 150 min (135–170). Median RAEPLND alone operative time was 38 min (32.75–41.25); for right and left side, 18 (15–29) and 20 min (15.75–30) (p=0.567). Median number of lymph nodes retrieved per patient was 25 (19.25–30); 13 (11–16) and 11 (8–15) for right and left side. In 19 patients (12.41%) metastasis was found at the level of pelvic lymph nodes. Median number of positive lymph nodes was 1 (1–4.6) per patient. Complications occurred in 11 patients (7.3%). Conclusions: the number of lymph nodes removed was comparable to published data about open series, allowing the increase of detection rate of lymph nodal metastasis for minimally invasive approach without compromising complications' rate if performing the procedure following reported technique.