Automated Identification of Key Steps in Robotic-Assisted Radical Prostatectomy Using Artificial Intelligence.

PURPOSE: The widespread use of minimally invasive surgery generates vast amounts of potentially useful data in the form of surgical video. However, raw video footage is often unstructured and unlabeled, thereby limiting its use. We developed a novel computer-vision algorithm for automated identification and labeling of surgical steps during robotic-assisted radical prostatectomy (RARP). MATERIALS AND METHODS: Surgical videos from RARP were manually annotated by a team of image annotators under the supervision of 2 urologic oncologists. Full-length surgical videos were labeled to identify all steps of surgery. These manually annotated videos were then utilized to train a computer vision algorithm to perform automated video annotation of RARP surgical video. Accuracy of automated video annotation was determined by comparing to manual human annotations as the reference standard. RESULTS: A total of 474 full-length RARP videos (median 149 minutes; IQR 81 minutes) were manually annotated with surgical steps. Of these, 292 cases served as a training dataset for algorithm development, 69 cases were used for internal validation, and 113 were used as a separate testing cohort for evaluating algorithm accuracy. Concordance between artificial intelligence‒enabled automated video analysis and manual human video annotation was 92.8%. Algorithm accuracy was highest for the vesicourethral anastomosis step (97.3%) and lowest for the final inspection and extraction step (76.8%). CONCLUSIONS: We developed a fully automated artificial intelligence tool for annotation of RARP surgical video. Automated surgical video analysis has immediate practical applications in surgeon video review, surgical training and education, quality and safety benchmarking, medical billing and documentation, and operating room logistics.

Mayo Clinic Validation of the AUA Risk Groups for Localized Renal Cell Carcinoma.

PURPOSE: The AUA guidelines introduced a new risk group stratification system based primarily on tumor stage and grade to guide surveillance for patients treated surgically for localized renal cell carcinoma (RCC). We sought to evaluate the predictive ability of these risk groups using progression-free survival (PFS) and cancer-specific survival (CSS), and to compare their performance to that of our published institutional risk models. MATERIALS AND METHODS: We queried our Nephrectomy Registry to identify adults treated with radical or partial nephrectomy for unilateral, M0, clear cell RCC, or papillary RCC from 1980 to 2012. The AUA stratification does not apply to other RCC subtypes as tumor grading for other RCC, such as chromophobe, is not routinely performed. PFS and CSS were estimated using the Kaplan-Meier method. Predictive abilities were evaluated using C indexes from Cox proportional hazards regression models. RESULTS: A total of 3191 patients with clear cell RCC and 633 patients with papillary RCC were included. For patients with clear cell RCC, C indexes for the AUA risk groups and our model were 0.780 and 0.815, respectively (P < .001) for PFS, and 0.811 and 0.857, respectively (P < .001), for CSS. For patients with papillary RCC, C indexes for the AUA risk groups and our model were 0.775 and 0.751, respectively (P = .002) for PFS, and 0.830 and 0.803, respectively (P = .2) for CSS. CONCLUSIONS: The AUA stratification is a parsimonious system for categorizing RCC that provides C indexes of about 0.80 for PFS and CSS following surgery for localized clear cell and papillary RCC.

End-Stage Kidney Disease After Partial and Radical Nephrectomy Among Patients with Severe Chronic Kidney Disease.

PURPOSE: AUA guidelines prioritize nephron sparing in patients with preexisting chronic kidney disease (CKD). However, few studies analyze long-term renal function in patients with preoperative severe CKD who undergo extirpative renal surgery. Herein, we compare the hazard of progression to end-stage kidney disease (ESKD) following partial nephrectomy (PN) and radical nephrectomy (RN) among patients with preoperative severe CKD. MATERIALS AND METHODS: Patients with stage 4 CKD who underwent PN or RN from 1970 to 2018 were identified. A multivariable Fine-Gray subdistribution hazard model was employed to assess associations with progression to ESKD accounting for the competing risk of death. RESULTS: A total of 186 patients with stage 4 CKD underwent PN (n = 71; 38%) or RN (n = 115; 62%) for renal neoplasms with median follow-up of 6.9 years (interquartile range 3.8-14.1). On multivariable analyses adjusting for competing risk of death, the subdistribution hazard ratio (SHR) for older age at surgery (SHR for 5-year increase 0.81; 95% CI 0.73-0.91; P < .001) and higher preoperative estimated glomerular filtration rate (SHR for 5-unit increase 0.63; 95% CI 0.47-0.84; P = .002) was associated with lower hazard of progression to ESKD. There was no significant difference in hazard of ESKD between PN and RN (SHR 0.82; 95% CI 0.50-1.33; P = .4). CONCLUSIONS: Among patients with preoperative severe CKD, higher preoperative estimated glomerular filtration rate was associated with lower hazard of progression to ESKD after extirpative surgery for renal neoplasms. We did not observe a significant difference in overall hazard for developing ESKD between PN and RN.

The optimal number of induction chemotherapy cycles in clinically lymph node-positive bladder cancer.

OBJECTIVE: To investigate the optimal number of induction chemotherapy cycles needed to achieve a pathological response in patients with clinically lymph node-positive (cN+) bladder cancer (BCa) who received three or four cycles of induction chemotherapy followed by consolidative radical cystectomy (RC) with pelvic lymph node dissection. PATIENTS AND METHODS: We included 388 patients who received three or four cycles of cisplatin/gemcitabine or (dose-dense) methotrexate, vinblastine, doxorubicin, and cisplatin (MVAC), followed by consolidative RC for cTanyN1-3M0 BCa. We compared pathological complete (pCR = ypT0N0) and objective response (pOR = yp ≤T1N0) between treatment groups. Predictors of pCR and/or pOR were assessed using uni- and multivariable logistic regression analysis. The secondary endpoints were overall (OS) and cancer-specific survival (CSS). We evaluated the association between the number of induction chemotherapy cycles administered and survival outcomes on multivariable Cox regression. RESULTS: Overall, 101 and 287 patients received three or four cycles of induction chemotherapy, respectively. Of these, 72 (19%) and 128 (33%) achieved pCR and pOR response, respectively. The pCR (20%, 18%) and pOR (40%, 31%) rates did not differ significantly between patients receiving three or four cycles (P > 0.05). The number of cycles was not associated with pCR or pOR on multivariable logistic regression analyses. The 2-year OS estimates were 63% (95% confidence interval [CI] 0.53-0.74) and 63% (95% CI 0.58-0.7) for patients receiving three or four cycles, respectively. Receiving three vs four cycles was not associated with OS and CSS on uni- or multivariable Cox regression analyses. CONCLUSION: Pathological response and survival outcomes did not differ between administering three or four induction chemotherapy cycles in patients with cN+ BCa. A fewer cycles (minimum three) may be oncologically sufficient in patients with cN+ BCa, while decreasing the wait for definitive local therapy in those patients who end up without a response to chemotherapy. This warrants further validation.

Cost-effectiveness of Adjuvant Pembrolizumab After Nephrectomy for High-risk Renal Cell Carcinoma: Insights for Patient Selection From a Markov Model.

PURPOSE: The KEYNOTE-564 trial demonstrated that adjuvant pembrolizumab after nephrectomy for clear cell renal cell carcinoma decreased the risk of disease progression and potentially overall mortality as well. Herein, we used a Markov model to weigh the costs, toxicities, and efficacy of pembrolizumab to further investigate its utility. MATERIALS AND METHODS: Decision-analytic Markov modeling was used to conduct a cost-utility analysis of adjuvant pembrolizumab versus observation after nephrectomy for high-risk clear cell renal cell carcinoma, using data from KEYNOTE-564 to inform model probabilities. Primary outcomes were quality-adjusted life years, Medicare costs, and incremental cost-effectiveness ratios. The willingness-to-pay threshold utilized was $100,000/quality-adjusted life year. RESULTS: At 5 years, adjuvant treatment with pembrolizumab resulted in 0.3 additional quality-adjusted life years at an additional cost of $99,484 relative to observation. Pembrolizumab was found not to be cost-effective at a 5-year time horizon (incremental cost-effectiveness ratio=$326,534). On sensitivity analysis, pembrolizumab became cost-effective if its per cycle cost was <$5,064 (base=$10,278) or its 5-year progression benefit was >18.8% (base 9%). Upon simulation, pembrolizumab was cost-effective for 29% of patients at 5 years. Specifically, we found that pembrolizumab would be cost-effective at 5 years for patients with at least a 59% 5 year risk of progression, which corresponds to a Mayo Progression-free Survival Score ≥10. CONCLUSIONS: At current prices, adjuvant pembrolizumab was found to be cost-effective only for the highest risk subset of clear cell renal cell carcinoma patients 5 years after treatment, including patients with complete metastasectomy, regional lymph node involvement, or ≥7cm pT3 tumors with sarcomatoid features. Longer-term trial data, including overall survival results, are necessary to confirm these extrapolations.

Effect of Dataset Size and Medical Image Modality on Convolutional Neural Network Model Performance for Automated Segmentation: A CT and MR Renal Tumor Imaging Study.

The aim of this study is to investigate the use of an exponential-plateau model to determine the required training dataset size that yields the maximum medical image segmentation performance. CT and MR images of patients with renal tumors acquired between 1997 and 2017 were retrospectively collected from our nephrectomy registry. Modality-based datasets of 50, 100, 150, 200, 250, and 300 images were assembled to train models with an 80-20 training-validation split evaluated against 50 randomly held out test set images. A third experiment using the KiTS21 dataset was also used to explore the effects of different model architectures. Exponential-plateau models were used to establish the relationship of dataset size to model generalizability performance. For segmenting non-neoplastic kidney regions on CT and MR imaging, our model yielded test Dice score plateaus of [Formula: see text] and [Formula: see text] with the number of training-validation images needed to reach the plateaus of 54 and 122, respectively. For segmenting CT and MR tumor regions, we modeled a test Dice score plateau of [Formula: see text] and [Formula: see text], with 125 and 389 training-validation images needed to reach the plateaus. For the KiTS21 dataset, the best Dice score plateaus for nn-UNet 2D and 3D architectures were [Formula: see text] and [Formula: see text] with number to reach performance plateau of 177 and 440. Our research validates that differing imaging modalities, target structures, and model architectures all affect the amount of training images required to reach a performance plateau. The modeling approach we developed will help future researchers determine for their experiments when additional training-validation images will likely not further improve model performance.

Role of Lymphadenectomy during Radical Cystectomy for Nonmuscle-Invasive Bladder Cancer: Results from a Multi-Institutional Experience.

PURPOSE: While lymph node dissection (LND) at radical cystectomy (RC) for muscle-invasive bladder cancer has been studied extensively, the role of LND for nonmuscle-invasive bladder cancer (NMIBC) remains incompletely defined. Herein, we aim to assess the association between extent of LND during RC for NMIBC and local pelvic recurrence-free survival (LPRS), cancer-specific survival (CSS) and overall survival (OS). MATERIALS AND METHODS: A multi-institutional retrospective review was performed of patients with NMIBC undergoing RC at 3 large tertiary referral centers. To identify a threshold for lymph node yield (LNY) to optimize LPRS, CSS and OS, separate Cox regression models were developed for each possible LNY threshold. Model performance including Q-statistics and hazard ratios (HRs) were used to identify optimal LNY thresholds. RESULTS: A total of 1,647 patients underwent RC for NMIBC, with a median LNY of 15 (quartiles 9,23). Model performance curves suggested LNY of 10 and 20 to optimize LPRS and CSS/OS, respectively. On multivariable regression, LNY >10 was associated with lower risk of LPR compared to LNY ≤10 (HR 0.63, 95% CI 0.42-0.93, p=0.02). Similarly, LNY >20 was associated with improved CSS (HR 0.67, 95% CI 0.52-0.87, p=0.002) and OS (HR 0.75, 95% CI 0.64-0.88, p <0.001) compared to LNY ≤20. Similar results were observed in the cT1 and cTis subgroups. CONCLUSIONS: Greater extent of LND during RC for NMIBC is associated with improved LPRS, CSS and OS, supporting the inclusion of LND during RC for NMIBC, particularly among patients with cTis or cT1 disease. Future prospective studies are warranted to assess the ideal anatomical template of LND in NMIBC.

A Contemporary Analysis of Urethral Recurrence following Radical Cystectomy.

PURPOSE: Oncologic outcomes following urethral recurrence (UR) remain incompletely described, with reports limited by small cohort sizes. We evaluated risk factors for UR as well as cancer-specific survival (CSS) and overall survival (OS) among patients with UR. MATERIALS AND METHODS: We reviewed our institutional radical cystectomy (RC) registry to identify patients with UR. Cox proportional hazards regression was used to assess risk factors for UR. Kaplan-Meier and Cox models were used to assess the relationship between UR and CSS/OS as well as to compare outcomes following symptomatic vs asymptomatic presentation of UR. RESULTS: Overall, 2,930 patients underwent RC from 1980 to 2018, with a median postoperative followup of 7.1 years (IQR 2.8-13.1), of whom 144 (4.9%) were subsequently diagnosed with UR. Carcinoma in situ (HR 1.98, 95% CI 1.30-3.04), multifocal disease (HR 1.59, 95% CI 1.07-2.36) and prostatic urethral involvement at RC (HR 3.01, 95% CI 1.98-4.57) were associated with increased risk of UR. UR was associated with decreased CSS (HR 7.30, 95% CI 5.46-9.76) and OS (HR 1.86, 95% CI 1.54-2.24). A total of 63/144 patients were diagnosed with UR based on symptoms, while 104/144 patients with UR underwent urethrectomy. Patients with symptomatic UR had higher tumor stage at urethrectomy (≥pT2 in 13.1% vs 3.1%, p=0.007), while patients with asymptomatic UR experienced longer median CSS (12.1 vs 6.1 years) and OS (8.30 vs 4.82 years; p=0.05 for both). CONCLUSIONS: We identified pathological risk factors for UR after RC and report adverse subsequent survival outcomes for these patients. Presentation with symptomatic UR was associated with higher tumor stage and poorer prognosis, supporting a value to continued urethral surveillance after RC.

Contemporary management of advanced prostate cancer: an evolving landscape.

Recent population-based studies suggest that the incidence of advanced and metastatic prostate cancer may be increasing. Concurrently with this apparent stage migration toward advanced disease, several major developments have occurred in the treatment paradigm for men with advanced prostate cancer. These include the US Food and Drug Administration approval of 8 novel agents over the last decade. In addition to novel pharmaceuticals, rapidly evolving diagnostic tools have emerged. This review provides a primer for clinicians who treat men with advanced prostate cancer, including medical oncologists, radiation oncologists, and urologists.

Large Single Institution Comparison of Perioperative Outcomes and Complications of Open Radical Cystectomy, Intracorporeal Robot-Assisted Radical Cystectomy and Robotic Extracorporeal Approach.

PURPOSE: Bladder cancer management options include open radical cystectomy and robot-assisted radical cystectomy with intracorporeal or extracorporeal urinary diversion. The existing literature shows no difference in the major complication rate between open radical cystectomy and extracorporeal urinary diversion. However, the emerging popularity of intracorporeal urinary diversion has exposed the need to compare a completely intracorporeal method to alternative approaches. To our knowledge the robotic intracorporeal advantage regarding major complications has not yet been established in an evaluation of all 3 modalities. We compared outcomes and complications of open, intracorporeal and extracorporeal cystectomy techniques at a high volume institution. MATERIALS AND METHODS: We queried a prospectively maintained database for patients who underwent radical cystectomy from 2011 to 2018 for an oncologic indication. Perioperative and pathological outcomes, and 30 and 90-day major complications were assessed. Statistical analyses were done using the Pearson chi-square, Kruskal-Wallis and Kaplan-Meier tests, and multivariable regression. RESULTS: A total of 948 patients met the study criteria, including 272, 301 and 375 treated with open radical cystectomy, intracorporeal urinary diversion and extracorporeal urinary diversion, respectively. Median followup was 26 months. Intracorporeal urinary diversion cases had lower estimated blood loss (p <0.001), shorter hospitalization (p <0.001) and a lower ileus rate (p=0.023) than extracorporeal urinary diversion and open radical cystectomy cases. Importantly, intracorporeal urinary diversion was associated with lower 30 and 90-day major complication rates vs extracorporeal urinary diversion and open radical cystectomy (90-day Clavien-Dindo III-V 16.9% vs 24.8% and 26.1%, respectively, p=0.015). There was no significant difference in the readmission rate according to the surgical approach. Multivariable predictors of increased 90-day major complications were patient age, the Charlson Comorbidity Index and operative time. On multivariable analysis intracorporeal urinary diversion was associated with reduced 90-day major complications (OR 0.58, p=0.037). CONCLUSIONS: In a 3-way comparison intracorporeal urinary diversion demonstrated a lower major complication rate and perioperative benefits compared to extracorporeal urinary diversion and open radical cystectomy.

Hospital Quality Metrics for Radical Cystectomy: Disease Specific and Correlated to Mortality Outcomes.

PURPOSE: Measuring quality is a high priority for health care systems globally. Despite the high perioperative morbidity, mortality, expenditures and performance variation of radical cystectomy there is a paucity of validated bladder cancer quality metrics. We aimed to create a hospital quality scoring system for radical cystectomy which is disease specific and associated with patient centered outcomes. MATERIALS AND METHODS: We used the National Cancer Database to identify hospitals where radical cystectomy was performed from 2004 to 2014. Mixed effects models were constructed to assess variation in hospital performance across 7 quality indicators. Indirect standardization was used to case mix adjust hospital performance. We assessed associations between quality indicators as well as the novel BC-QS (Bladder Cancer Quality Score) composite hospital quality metric with 30-day, 90-day and overall mortality using logistic and Cox regression, respectively. RESULTS: At 1,200 facilities radical cystectomy was performed in a total of 48,341 patients from 2004 to 2014. Mixed effects models demonstrated significant between hospital variation across all quality indicators after case mix adjustment. The composite BC-QS metric was composed of the hospital positive margin rate, the lymph node dissection rate and the neoadjuvant chemotherapy rate. Better BC-QS performance was associated with lower 30-day and 90-day mortality (adjusted OR 0.78, 95% CI 0.64-0.96, and OR 0.84, 95% CI 0.72-0.97, respectively) and overall mortality (HR 0.86, 95% CI 0.81-0.92). Hospitals with a higher BC-QS had higher volume and more were affiliated with an academic institution than hospitals with a lower BC-QS (p <0.0001). CONCLUSIONS: The BC-QS captures variations in the hospital performance of radical cystectomy and it shows an association of higher quality with lower patient mortality. Our validation of this quality metric provides support for its potential use by policy makers and payers in efforts to measure hospital quality for high cost surgeries.

Bladder-sparing treatment of nonmetastatic muscle-invasive bladder cancer.

Bladder-sparing therapies for the treatment of nonmetastatic muscle-invasive bladder cancers are included in both American and European guidelines. Numerous treatment approaches have been described, including partial cystectomy, radiation monotherapy, and radical transurethral resection. However, the most oncologically favorable and well-studied regimen employs a multimodal approach that consists of maximal transurethral resection of the bladder tumor followed by concurrent radiosensitizing chemotherapy and radiotherapy. This sequence, referred to as trimodal therapy (TMT), has been evaluated with robust retrospective comparative studies and prospective series, although a randomized trial comparing TMT with radical cystectomy has not been performed. Despite promising reports of 5-year overall survival rates of 50% to 70% in well-selected patients, relatively few patients qualify as ideal candidates for TMT. Specifically, contemporary series exclude patients who have clinical stage T3 disease, multifocal tumors, coexisting carcinoma in situ, or hydronephrosis. Herein, we review all forms of bladder-preserving therapies with an emphasis on TMT, highlighting the rationale of each component, survival outcomes, and future directions.

Contemporary management of advanced renal cell carcinoma.

Kidney cancer is the eighth most commonly diagnosed cancer in the United States, and nearly one-third of patients have locally advanced or metastatic disease at presentation. Historically, survival outcomes for patients with advanced disease have been poor. In recent years, several novel targeted agents have emerged for the management of advanced renal cell carcinoma that have changed treatment paradigms. At the same time, surgical therapy continues to have a critical role in the management of selected patients. Recent medical and surgical advances have improved the prognosis for patients with a diagnosis of advanced disease. This review provides an overview of the current treatment landscape for patients with advanced renal cell carcinoma.

Testicular seminoma: oncologic rationale and role of surgery in treatment.

Seminomas account for approximately 50% of all cases of testicular cancer. Testicular cancer is a highly curable disease that can be broadly classified as either seminomatous or nonseminomatous; the management and treatment of the 2 forms vary widely. Although surgery plays a large role in the management of nonseminoma, its role in the management of seminoma is much more limited. Most clinicians in the United States choose orchiectomy followed by surveillance for patients with stage I seminomatous disease, and chemotherapy or radiation-followed by surgery for the management of residual masses-for patients with disease that is stage II and higher. Recently, clinicians have proposed a larger role for surgery in stage II seminoma to avoid the long-term toxic effects of chemotherapy and radiation therapy. In this review, we discuss the oncologic rationale for the treatment of seminoma, the role of surgery, and the use of minimally invasive operative techniques for retroperitoneal lymph node dissection.

Testosterone Therapy after Radiation Therapy for Low, Intermediate and High Risk Prostate Cancer.

PURPOSE: Limited literature exists regarding the safety of testosterone therapy in men treated for prostate cancer. We present multi-institutional data on testosterone therapy in hypogonadal men with prostate cancer treated with radiation therapy. MATERIALS AND METHODS: We retrospectively reviewed the records of hypogonadal men treated with testosterone therapy after radiation therapy for prostate cancer at 4 institutions. Serum testosterone, free testosterone, estradiol, sex hormone-binding globulin, prostate specific antigen, prostate specific antigen velocity and prostate biopsy findings were analyzed. RESULTS: A total of 98 men were treated with radiation therapy. Median age was 70.0 years (range 63.0 to 74.3) at initiation of testosterone therapy. Median baseline testosterone was 209 ng/dl (range 152 to 263) and median baseline prostate specific antigen was 0.08 ng/ml (range 0.00 to 0.33). In the cohort the tumor Gleason score was 5 in 3 men (3.1%), 6 in 44 (44.9%), 7 in 28 (28.6%), 8 in 7 (7.1%) and 9 in 4 (4.1%). Median followup was 40.8 months (range 1.5 to 147). Serum testosterone increased to a median of 420 ng/dl (range 231 to 711) during followup (p <0.001). Overall a nonsignificant increase in mean prostate specific antigen was observed from 0.08 ng/ml at baseline to 0.09 ng/ml (p = 0.05). Among patients at high risk prostate specific antigen increased from 0.10 to 0.36 ng/ml (p = 0.018). Six men (6.1%) met criteria for biochemical recurrence. CONCLUSIONS: Testosterone therapy in men following radiation therapy for prostate cancer was associated with a minor increase in serum prostate specific antigen and a low rate of biochemical recurrence.

The significance of sperm heads and tails within the vasal fluid during vasectomy reversal.

INTRODUCTION: The finding of only sperm heads and/or short tails (SHST) during vasectomy reversal (VR) creates a difficult decision for the best method of vasal reconstruction, i.e. vasovasostomy (VV) or epididymovasostomy (EV). Using outcome analyses, we report the impact of SHST alone and combined with qualitative analysis of gross fluid quality in predicting successful VR. MATERIALS AND METHODS: The records of 356 men who underwent VR by a single surgeon from 2005 to 2012 were retrospectively reviewed. Intravasal fluid was assessed for gross quality (i.e., clear, opaque, pasty or creamy) as well as microscopic composition (i.e., motile or non-motile whole sperm, SHST or no sperm). The post-operative patency rates and semen analysis parameters were assessed. RESULTS: Fourteen men (3.9%) demonstrated SHST bilaterally in the vasal fluid. The median duration from vasectomy was 6.0 years (interquartile range 4.0-9.8). Bilateral VVs were performed on 12 men (86%), while two men (14%) had a unilateral VV and a contralateral EV. Of the 26 vasa undergoing VR, the majority of the fluid quality was classified as creamy (n = 20 vasa, 76.9%). The remaining fluid was classified as pasty (n = 3 vasa, 11.5%), opaque (n = 2 vasa, 7.7%) and clear (n = 1 vasa, 3.8%). In cases undergoing bilateral VV with only SHST, patency rates were 90.9%, and both cases of unilateral EV were patent (100%). CONCLUSIONS: VV was successful in 90.9% of patients undergoing VR in the setting of SHST alone. Even when creamy or pasty fluid was present, the results surpassed the expected patency rate for an EV. Therefore, the presence of only SHST, regardless of fluid quality, should not dissuade the surgeon from performing a VV. CONCLUSIONS: VV was successful in 90.9% of patients undergoing VR in the setting of SHST alone. Even when creamy or pasty fluid was present, the results surpassed the expected patency rate for an EV. Therefore, the presence of only SHST, regardless of fluid quality, should not dissuade the surgeon from performing a VV.

Navigating Now and Next: Recent Advances and Future Horizons in Robotic Radical Prostatectomy.

Robotic-assisted radical prostatectomy (RARP) has become the leading approach for radical prostatectomy driven by innovations aimed at improving functional and oncological outcomes. The initial advancement in this field was transperitoneal multiport robotics, which has since undergone numerous technical modifications. These enhancements include the development of extraperitoneal, transperineal, and transvesical approaches to radical prostatectomy, greatly facilitated by the advent of the Single Port (SP) robot. This review offers a comprehensive analysis of these evolving techniques and their impact on RARP. Additionally, we explore the transformative role of artificial intelligence (AI) in digitizing robotic prostatectomy. AI advancements, particularly in automated surgical video analysis using computer vision technology, are unprecedented in their scope. These developments hold the potential to revolutionize surgeon feedback and assessment and transform surgical documentation, and they could lay the groundwork for real-time AI decision support during surgical procedures in the future. Furthermore, we discuss future robotic platforms and their potential to further enhance the field of RARP. Overall, the field of minimally invasive radical prostatectomy for prostate cancer has been an incubator of innovation over the last two decades. This review focuses on some recent developments in robotic prostatectomy, provides an overview of the next frontier in AI innovation during prostate cancer surgery, and highlights novel robotic platforms that may play an increasing role in prostate cancer surgery in the future.

Automated surgical step recognition in transurethral bladder tumor resection using artificial intelligence: transfer learning across surgical modalities.

Objective: Automated surgical step recognition (SSR) using AI has been a catalyst in the "digitization" of surgery. However, progress has been limited to laparoscopy, with relatively few SSR tools in endoscopic surgery. This study aimed to create a SSR model for transurethral resection of bladder tumors (TURBT), leveraging a novel application of transfer learning to reduce video dataset requirements. Materials and methods: Retrospective surgical videos of TURBT were manually annotated with the following steps of surgery: primary endoscopic evaluation, resection of bladder tumor, and surface coagulation. Manually annotated videos were then utilized to train a novel AI computer vision algorithm to perform automated video annotation of TURBT surgical video, utilizing a transfer-learning technique to pre-train on laparoscopic procedures. Accuracy of AI SSR was determined by comparison to human annotations as the reference standard. Results: A total of 300 full-length TURBT videos (median 23.96 min; IQR 14.13-41.31 min) were manually annotated with sequential steps of surgery. One hundred and seventy-nine videos served as a training dataset for algorithm development, 44 for internal validation, and 77 as a separate test cohort for evaluating algorithm accuracy. Overall accuracy of AI video analysis was 89.6%. Model accuracy was highest for the primary endoscopic evaluation step (98.2%) and lowest for the surface coagulation step (82.7%). Conclusion: We developed a fully automated computer vision algorithm for high-accuracy annotation of TURBT surgical videos. This represents the first application of transfer-learning from laparoscopy-based computer vision models into surgical endoscopy, demonstrating the promise of this approach in adapting to new procedure types.