Artificial intelligence is highly regarded as the most promising future technology that will have a great impact on healthcare across all specialties. Its subsets, machine learning, deep learning, and artificial neural networks, are able to automatically learn from massive amounts of data and can improve the prediction algorithms to enhance their performance. This area is still under development, but the latest evidence shows great potential in the diagnosis, prognosis, and treatment of urological diseases, including bladder cancer, which are currently using old prediction tools and historical nomograms. This review focuses on highly significant and comprehensive literature evidence of artificial intelligence in the management of bladder cancer and investigates the near introduction in clinical practice.

Introduction To evaluate the efficacy and safety of ureteroscopic thulium laser treatment of upper urinary tract carcinoma (UTUC). Material and Methods Forty-two consecutive patients underwent conservative thulium laser treatment for UTUC at two referral institutions. All patients underwent preliminary biopsy and then laser vaporization. A 272 μm and a 365 μm laser fibers were used with a flexible and semirigid scope, respectively. Ablation was carried out with a 10-20 Watt power. Results Mean age at surgery was 68 years old (SD 10.7). Mean tumour size was 14.3 (range 2-30) mm. Preliminary biopsy revealed the presence of low-grade disease in 29 (69.1%) patients, high-grade in 4 (9.5%) and one carcinoma in situ 1 (2.4%), while it was not conclusive in 8 (19%) cases. Final stage was pTa and pTis in 41 (97.6%) and 1 (2.4%) patients, respectively. The 38% (16) experienced Clavien-Dindo grade I complication, 47.6% (20) a grade II and 2.4% (1) a grade III. Five (12%) patients underwent a second-look procedure due to residual disease. Eight (19%) patients experienced clinical recurrence. The median estimated recurrence-free survival was 44 months (SE 3.68). Four patients (9.5%) underwent a nephroureterectomy. Final pathological stage was pTis, pT3 high grade, pTa low grade and pT0. Median follow-up was 26.3 (range 2-54) months, and no progression or upstaging of disease occurred. Conclusions Thulium laser management of UTUC is a safe and efficacious conservative treatment. Our experience shows optimal vaporization and haemostatic control in the absence of major complications.

Background: The role of robot assistance is increasingly gaining importance among all major surgical uro-oncological procedures (MSUPs). However, contemporary analyses showed that total hospital charges (THCGs) related to robot-assisted procedures exceed those of open procedures. Based on increasing familiarity with robot-assisted surgery, we postulated that THCGs may have decreased over the past half-decade. Thus, we tested contemporary trends and THCGs related to robot-assisted vs nonrobot-assisted MSUPs. Materials and Methods: Within the National Inpatient Sample database (2009-2015), we identified patients who underwent robot-assisted vs nonrobot-assisted (open or laparoscopic) MSUPs, which included radical prostatectomy (RP), radical nephrectomy (RN), partial nephrectomy (PN), and radical cystectomy (RC). Rates of robot-assisted MSUPs were evaluated using estimated annual percentage changes (EAPCs) analyses. The t-test was used to examine statistically significant differences between mean THCGs according to either robot-assisted or nonrobot-assisted approach. Finally, linear regression analyses were tested for annual variation in the mean THCGs. Results: Of 128,367 MSUPs, 47.7% were robot-assisted. Overall, robot-assisted surgery rates among MSUPs increased from 40.3% to 57.6% (EAPC: +6.3%, p < 0.001) between 2009 and 2015. The mean THCGs for robot-assisted RP, RN, PN, and RC were $13,799, $18,789, $16,574, and $33,575, respectively. The observed mean THCGs differences between robot-assisted and nonrobot-assisted MSUPs were +$1594, +$1592, and +$1829 for RP, RN, and RC, respectively (all p < 0.05). Conversely, no statistically significant difference in the mean THCGs was reported between robot-assisted and nonrobot-assisted PN (+$367, p > 0.05). Finally, the annual observed mean THCGs linearly decreased for all robot-assisted MSUPs during the study period. Conclusions: Rates of robot-assisted MSUPs exponentially increased between 2009 and 2015. Although the mean THCGs decreased in a significant manner during the study period for all MSUPs, THCGs of robot-assisted RP, RN, and RC still exceed those of their respective nonrobot-assisted counterparts. Conversely, no differences in the mean THCGs were reported between robot-assisted vs nonrobot-assisted PN.

Background: Several biochemical and clinical markers have been proposed for selecting patients for active surveillance (AS). However, some of these are expensive and not easily accessible. Moreover, currently about 30% of patients on AS harbor aggressive disease. Hence, there is an urgent need for other tools to accurately identify patients with low-risk prostate cancer (PCa). Patients: We retrospectively reviewed the medical records of 260 patients who underwent radical prostatectomy and were eligible for AS according to the following criteria: clinical stage T2a or less, prostate-specific antigen level < 10 ng/mL, 2 or fewer cores involved with cancer, Gleason score (GS) ≤6 grade, and prostate-specific antigen density < 0.2 ng/mL/cc. Methods: Univariate and multivariate analyses were performed to evaluate the association of patient and tumor characteristics with reclassification, defined as upstaged (pathological stage >pT2) and upgraded (GS ≥7) disease. A base model (age, prostate-specific antigen, prostate volume, and clinical stage) was compared with models considering neutrophil to lymphocyte ratio (NLR) or platelets to lymphocyte ratio (PLR), monocyte to lymphocyte (MLR), and eosinophil to lymphocyte ratio (ELR). OR and 95% CI were calculated. Finally, a decision curve analysis was performed. Results: Univariate and multivariate analyses showed that NLR, PLR, and ELR upgrading were significantly associated with upgrading (ORs ranging from 2.13 to 4.13), but not with upstaging except for MLR in multivariate analysis, showing a protective effect. Conclusion: Our results showed that NLR, PLR, and ELR are predictors of Gleason upgrading. Therefore, these inexpensive and easily available tests might be useful in the assessment of low-risk PCa, when considering patients for AS.