Privacy-proof Live Surgery Streaming: Development and Validation of a Low-cost, Real-time Robotic Surgery Anonymization Algorithm.

OBJECTIVE: Develop a pioneer surgical anonymization algorithm for reliable and accurate real-time removal of out-of-body images validated across various robotic platforms. BACKGROUND: The use of surgical video data has become a common practice in enhancing research and training. Video sharing requires complete anonymization, which, in the case of endoscopic surgery, entails the removal of all nonsurgical video frames where the endoscope can record the patient or operating room staff. To date, no openly available algorithmic solution for surgical anonymization offers reliable real-time anonymization for video streaming, which is also robotic-platform and procedure-independent. METHODS: A data set of 63 surgical videos of 6 procedures performed on four robotic systems was annotated for out-of-body sequences. The resulting 496.828 images were used to develop a deep learning algorithm that automatically detected out-of-body frames. Our solution was subsequently benchmarked against existing anonymization methods. In addition, we offer a postprocessing step to enhance the performance and test a low-cost setup for real-time anonymization during live surgery streaming. RESULTS: Framewise anonymization yielded a receiver operating characteristic area under the curve score of 99.46% on unseen procedures, increasing to 99.89% after postprocessing. Our Robotic Anonymization Network outperforms previous state-of-the-art algorithms, even on unseen procedural types, despite the fact that alternative solutions are explicitly trained using these procedures. CONCLUSIONS: Our deep learning model, Robotic Anonymization Network, offers reliable, accurate, and safe real-time anonymization during complex and lengthy surgical procedures regardless of the robotic platform. The model can be used in real time for surgical live streaming and is openly available.

Complications of single-port robot-assisted radical prostatectomy: multi-institutional analysis from the Single-Port Advanced Research Consortium (SPARC).

OBJECTIVE: To evaluate the perioperative complications of single-port robot-assisted radical prostatectomy (SP-RARP). PATIENTS AND METHODS: A retrospective review was performed on the prospectively maintained, Institutional Review Board-approved, multi-institutional Single-Port Advanced Research Consortium (SPARC) database. A total of 1103 patients were identified who underwent three different approaches of SP-RARP between 2019 and 2022 using the purpose-built SP robotic platform. In addition to baseline clinical, perioperative outcomes, this study comprehensively analysed for any evidence of intraoperative complication, as well as postoperative complication and readmission within 90 days of the respective surgery. RESULTS: Of the 244, 712, and 147 patients who underwent transperitoneal, extraperitoneal, and transvesical SP-RARP, respectively, intraoperative complications were noted in five patients (0.4%), all of which occurred during the transperitoneal approach. Two patients had bowel serosal tears, two had posterior button-holing of the bladder necessitating repair, and one patient had an obturator nerve injury. Postoperative complications were noted in 143 patients (13%) with major complications (Clavien-Dindo Grade ≥III) only identified in 3.7% of the total cohort. The most common complications were lymphocele (3.9%), acute urinary retention (2%), and urinary tract infection (1.9%). The 90-day re-admission rate was 3.9%. CONCLUSION: The SP-RARP is a safe and effective procedure with low complication and readmission rates regardless of the approach. These results are comparable to current multi-port RARP literature.

The Surgical Learning Curve for Biochemical Recurrence After Robot-assisted Radical Prostatectomy.

BACKGROUND: Improved cancer control with increasing surgical experience-the learning curve-was demonstrated for open and laparoscopic prostatectomy. In a prior single-center study, we found that this might not be the case for robot-assisted radical prostatectomy (RARP). OBJECTIVE: To investigate the relationship between prior experience of a surgeon and biochemical recurrence (BCR) after RARP. DESIGN, SETTING, AND PARTICIPANTS: We retrospectively analyzed the data of 8101 patients with prostate cancer treated with RARP by 46 surgeons at nine institutions between 2003 and 2021. Surgical experience was coded as the total number of robotic prostatectomies performed by the surgeon before the patient operation. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: We evaluated the relationship of prior surgeon experience with the probability of BCR adjusting for preoperative prostate-specific antigen, pathologic stage, grade, lymph-node involvement, and year of surgery. RESULTS AND LIMITATIONS: Overall, 1047 patients had BCR. The median follow-up for patients without BCR was 33 mo (interquartile range: 14, 61). After adjusting for case mix, the relationship between surgical experience and the risk of BCR after surgery was not statistically significant (p = 0.2). The 5-yr BCR-free survival rates for a patient treated by a surgeon with prior 10, 250, and 1000 procedures performed were, respectively, 82.0%, 82.7%, and 84.8% (absolute difference between 10 and 1000 prior procedures: 1.6% [95% confidence interval: 0.4%, 3.3%). Results were robust to a number of sensitivity analyses. CONCLUSIONS: These findings suggest that, as opposed to open and laparoscopic radical prostatectomy, surgeons performing RARP achieve adequate cancer control in the early phase of their career. Further research should explore why the learning curve for robotic surgery differs from prior findings for open and laparoscopic radical prostatectomy. We hypothesize that surgical education, including simulation training and the adoption of objective performance metrics, is an important mechanism for flattening the learning curve. PATIENT SUMMARY: We investigated the relationship between biochemical recurrence after robot-assisted radical prostatectomy and surgeon's experience. Surgeons at an early stage of their career had similar outcomes to those of more experienced surgeons, and we hypothesized that surgical education in robotics might be an important determinant of such a finding.

Low Risk of Postoperative Hernia Following Single-port Robot-assisted Radical Prostatectomy: A Report From the Single-port Advanced Research Consortium (SPARC).

OBJECTIVE: To evaluate the risk of postoperative hernia following different approaches of single-port robot-assisted radical prostatectomy (SP-RARP). METHODS: A retrospective review was performed on patients who underwent SP-RARP between February 2019 and December 2022. Demographic and clinical information was collected from the multi-institutional, prospectively-maintained Single-Port Advanced Research Consortium (SPARC) database. Data were analyzed using IBM Statistical Packaging for Social Sciences (SPSS) version 29.0 with descriptive statistics as presented. RESULTS: A total of 1103 patients were identified, consisting of 244 (22.1%), 712 (64.6%), and 147 (13.3%) cases performed via transperitoneal, extraperitoneal (EP), and transvesical (TV) approaches, respectively. During a median follow-up time of 11 months (interquartile range 5.7-17.1 months), only two cases of incisional hernia were reported. Both cases occurred following transperitoneal SP-RARP with one patient requiring surgical repair. There remains no evidence of postoperative hernia following EP and TV SP-RARP at the completion of our review. CONCLUSION: SP-RARP was associated with low risk for postoperative hernia. The risk was lower following TV and EP SP-RARP where the peritoneum is preserved.

The surgical learning curve for salvage robot-assisted radical prostatectomy: a prospective single-surgeon study.

BACKGROUND: The aim of this study was to report the overall results and the learning curve (LC) in salvage robot-assisted radical prostatectomy (sRARP) patients, in terms of morbidity, oncological and functional outcomes in a single surgeon tertiary-referral center. METHODS: One hundred and twenty patients underwent sRARP by a single surgeon (V.P.) from 2008 to 2018. To assess the trends in the learning experience they were sub-divided in 4 groups of 30 consecutive patients based on date of surgery. The Kaplan-Meier method and regression models were used to identify survival estimations and predictors of potency, continence and biochemical failure (BCF) at 12 months. RESULTS: As the learning experience for sRALP increased operative time (OT) was significantly shorter (from 139.5 to 121 minutes) and the amount of nerve-sparing (NS) undertaken increased (from 46% to 80%). While complications rate remained stable, estimated blood loss (EBL) and radiographic anastomotic leaks (RAL) decreased through the groups (from 124 to 69 ml and 40% to 16,7%, respectively). BCF and continence rates at 12 months after sRARP were similar among groups (23-36% and 36,7-50%, respectively) and chance of potency rates tended to increase (from 3.3% to 16-23%) but was not statistically significant. In a multivariate analysis, predictors for BCF were PSM and GS 8-10. Non-radiation primary treatment was the unique predictor of continence at 12 months after sRARP. CONCLUSIONS: Our data may suggest a decreasing trend in terms of OT and EBL through the sRARP learning curve. While morbidity remained stable through the time, RAL trended towards a decline. A higher degree of NS was observed through the groups and there was a slight correlation trend between surgical expertise and potency recovery. PSM and GS 8-10 were predictors of BCF and non-radiation primary treatment predicted a better continence after sRARP.