Estimating Surgical Urethral Length on Intraoperative Robot-Assisted Prostatectomy Images using Artificial Intelligence Anatomy Recognition.

Objective To construct a Convolutional Neural Network (CNN) model that can recognize and delineate anatomic structures on intraoperative video frames of robot-assisted radical prostatectomy (RARP) and to use these annotations to predict the surgical urethral length (SUL). Background Urethral dissection during RARP impacts patient urinary incontinence (UI) outcomes, and requires extensive training. Large differences exist between incontinence outcomes of different urologists and hospitals. Also, surgeon experience and education are critical towards optimal outcomes. Therefore new approaches are warranted. SUL is associated with UI. Artificial intelligence (AI) surgical image segmentation using a CNN could automate SUL estimation and contribute towards future AI-assisted RARP and surgeon guidance. Methods Eighty-eight intraoperative RARP videos between June 2009 and September 2014 were collected from a single center. 264 frames were annotated according to: prostate, urethra, ligated plexus and catheter. Thirty annotated images from different RARP videos were used as a test dataset. The Dice coefficient (DSC) and 95th percentile Hausdorff distance (Hd95) were used to determine model performance. SUL was calculated using the catheter as a reference. Results The DSC of the best performing model were 0.735 and 0.755 for the catheter and urethra classes respectively, with a Hd95 of 29.27 and 72.62 respectively. The model performed moderately on the ligated plexus and prostate. The predicted SUL showed a mean difference of 0.64 - 1.86mm difference versus human annotators, but with significant deviation (SD 3.28 - 3.56). Conclusion This study shows that an AI image segmentation model can predict vital structures during RARP urethral dissection with moderate to fair accuracy. SUL estimation derived from it showed large deviations and outliers when compared to human annotators, but with a very small mean difference (<2mm). This is a promising development for further research on AI-assisted RARP. Keywords Prostate cancer, Anatomy recognition, Artificial intelligence, Continence, Urethral length.

Reassessment of Prostate Biopsy Specimens for Patients Referred for Robot-assisted Radical Prostatectomy Rarely Influences Surgical Planning.

BACKGROUND: The minimum volume standard is 100 robot-assisted radical prostatectomy (RARP) procedures per hospital in the Netherlands, so patients have to be referred to high-volume surgical centers for RARP. During preoperative work-up, prostate biopsies taken elsewhere are reassessed, with upgrading or downgrading of the initial Gleason grade group a possible consequence. OBJECTIVE: To determine if prostate biopsy reassessment leads to adjustment of the surgical plan regarding a nerve-sparing approach and extended pelvic lymph node dissection (ePLND) during RARP. DESIGN SETTING AND PARTICIPANTS: For 125 men who were referred to the Prosper prostate center at Canisius Wilhelmina Hospital (CWH) in the Netherlands between 2013 and 2016, results for the initial assessment of prostate biopsy by a local uropathologist were compared to results for biopsy reassessment by dedicated uropathologists at CWH. RESULTS AND LIMITATIONS: The pathologists reached agreement in 80% of the cases. In cases for which there was disagreement (n = 25), biopsy revision involved upgrading of the initial grade group in 68% and downgrading in 32%. Biopsy reassessment led to a change in surgical plan in ten cases (8%). As a result of upgrading, ePLND was performed in three patients (2%). ePLND was omitted in one patient (1%) because of downgrading. For three patients (2%) a non-nerve-sparing procedure was planned after upgrading of the initial grade group. For four patients (3%), a unilateral nerve-sparing procedure was performed after downgrading. CONCLUSIONS: This study shows that there is large interobserver agreement between uropathologists in the assessment of Gleason grade group in prostate biopsy specimens. Reassessment rarely leads to a change in surgical plan regarding the indication for a nerve-sparing approach and ePLND. Therefore, reassessment of prostate biopsy before radical prostatectomy can be omitted when the initial pathological assessment was performed by a dedicated uropathologist. PATIENT SUMMARY: Reassessment of the initial prostate biopsy specimen for patients referred to a specialist center for robot-assisted removal of the prostate rarely influences surgical planning and can be omitted.

Diagnostic accuracy of 18F-fluciclovine PET/CT in primary lymph node staging of prostate cancer.

INTRODUCTION: To determine preoperative diagnostic accuracy of 18F-fluciclovine PET/CT-scan in detection (or exclusion) of lymph node metastases (LNM) in men with prostate cancer (PCa) in comparison to the histopathological results of the extended pelvic lymph node dissection (e-PLND). METHODS: A retrospective medical records-based cohort study, including 47 men with primary PCa who received 18F-fluciclovine PET/CT and subsequent e-PLND for lymph node staging. Incidence and number of visualized LNM, their locations and diameters on 18F-fluciclovine PET/CT were recorded in comparison to the histopathological results of the e-PLND as reference. Positive predictive value (PPV), negative predictive value (NPV), sensitivity, specificity and diagnostic accuracy of 18F-fluciclovine PET/CT were calculated on the basis of histopathology results after e-PLND. RESULTS: Forty-seven men were eligible for analysis. Median lymph node yield was 19 (range 10-70). A total of 996 lymph nodes were removed, and 59 metastases were found in 21 cases (45%). Preoperative PET was issued 'positive' in 11 men and in 9 of them (82%) this was histopathologically confirmed resulting in a PPV of 82% (95% CI, 51-96). On the contrary, PET was issued 'negative' in 36 cases, but in 12 of them (33%) metastases were detected in the e-PLND specimen, resulting in an NPV of 67% (95% CI, 50-80). The patient-based sensitivity was 43% (95% CI, 24-64) and the patient-based specificity rate was 92% (95% CI, 75-99), whereas overall diagnostic accuracy was established to be 70% in the present cohort. CONCLUSION: 18F-Fluciclovine PET/CT has a high specificity and positive predicted value for the presence of LNM in men with prostate cancer. However, the sensitivity and NPV seem to be limited to exclude the absence of LNM at a clinically acceptable level. Prospective evaluation is necessary to define patients who may benefit from 18F-fluciclovine PET/CT as a triage test for the indication for e-PLND.

Identifying the relationship between postoperative urinary continence and residual urethra stump measurements in robot assisted radical prostatectomy patients.

OBJECTIVE: To investigate the feasibility of urethral stump length and width measurements in recorded videos of robot assisted radical prostatectomy procedures using the Kinovea software and to assess if these measurements could be used as predictors of postoperative urinary continence. METHODS: Fifty-three patients were selected from an institutional database of 1400 cases and included in the study. All videos were analysed using the computer software 'Kinovea'. All measurements were performed using the inserted bladder catheter as a reference point. RESULTS: The reference point (bladder catheter) was available in 33 out of 53 patients. The median surgical urethral length (SUL) was significantly higher in the continent group (1050 vs. 1294 mm, p = 0.018). The urethral width measurements did not show a difference between the groups. In order to validate the Kinovea software as an accurate tool for the measurement of the urethral stump length and width results were correlated with the magnetic resonance imaging measurements of the urethra. CONCLUSIONS: The results of this study showed a significantly longer median SUL incontinent patients.

Analysis of the video motion tracking system "Kinovea" to assess surgical movements during robot-assisted radical prostatectomy.

BACKGROUNDS: Robot-assisted surgery facilitated the possibility to evaluate the surgeon's skills by recording and evaluating the robot surgical images. The aim of this study was to investigate the possibility of using a computer programme (Kinovea) for objective assessment of surgical movements in previously recorded in existing robot-assisted radical prostatectomy (RARP) videos. METHODS: Twelve entire RARP procedures were analysed by a trained researcher using the computer programme "Kinovea" to perform semi-automated assessment of surgical movements. RESULTS: Data analysis showed Kinovea was on average able to automatically assess only 22% of the total surgical duration per video of the robot-assisted surgery. On average, it lasted 4 hours of continued monitoring by the researcher to assess one RARP using Kinovea. CONCLUSION: Although we proved it is technically possible to use the Kinovea system in retrospective analysis of surgical movement in robot-assisted surgery, the acquired data do not give a comprehensive enough analysis of the video to be used in skills assessment.

A Prospective, Observational, Multicentre Study Concerning Nontechnical Skills in Robot-assisted Radical Cystectomy Versus Open Radical Cystectomy.

INTRODUCTION AND HYPOTHESES: valuation of surgical skills, both technical and nontechnical, is possible through observations and video analysis. Besides technical failures, adverse outcomes in surgery can also be related to hampered communication, moderate teamwork, lack of leadership, and loss of situational awareness. Even though some surgeons are convinced about nontechnical skills being an important part of their professionalisation, there is paucity of data about a possible relationship between nontechnical skills and surgical outcome. In robot-assisted surgery, the surgeon sits behind the console and is at a remote position from the surgical field and team, making communication more important than in open surgery and conventional laparoscopy. A lack of structured research makes it difficult to assess the value of the different analysis methods for nontechnical skills, particularly in robot-assisted surgery. Our hypothesis includes the following: (1) introduction of robot-assisted surgery leads to an initial decay in nontechnical skills behaviour during the learning curve of the team, (2) nontechnical skills behaviour is more explicitly expressed in experienced robot-assisted surgery teams than in experienced open surgery teams, and (3) introduction of robot-assisted surgery leads to the development of different forms of nontechnical skills behaviour compared with open surgery. DESIGN: This study is a prospective, observational, multicentre, nonrandomised, case-control study including bladder cancer patients undergoing either an open radical cystectomy or a robot-assisted radical cystectomy at the Catharina Hospital Eindhoven, the Netherlands, or at the Netherlands Cancer Institute, Antoni van Leeuwenhoek Hospital Amsterdam. All patients are eligible for inclusion; there are no exclusion criteria. The Catharina Hospital Eindhoven, the Netherlands, performs on average 35 radical cystectomies a year. The Netherlands Cancer Institute, Antoni van Leeuwenhoek Hospital Amsterdam, performs on average 100 radical cystectomies a year. PROTOCOL OVERVIEW: The choice of treatment is at the discretion of the patient and the surgeon. Patient results will be obtained prospectively. Pathology results as well as complications occurring within 90 d following surgery will be registered. Surgical complications will be registered according to the Clavien-Dindo system. MEASUREMENTS: Nontechnical skills will be observed using five different methods: (1) NOTSS: Nontechnical Skills for Surgeons; (2) Oxford NOTECHS II: a modified theatre team nontechnical skills scoring system; (3) OTAS: Observational Teamwork Assessment for Surgery; (4) Interpersonal and Cognitive Assessment for Robotic Surgery (ICARS): evaluation of nontechnical skills in robotic surgery; and (5) analysis of human factors. Technical skills in robot-assisted radical cystectomy will be analysed using two different methods: (1) GEARS: Global Evaluative Assessment of Robotic Skill and (2) GERT: Generic Error Rating Tool. SAFETY CRITERIA AND REPORTING: Formal ethical approval has been provided by Medical research Ethics Committees United (MEC-U), The Netherlands (reference number W19.048). We hope to present the results of this study to the scientific community at conferences and in peer-reviewed journals. STATISTICAL ANALYSIS: Frequency statistics will be calculated for patient demographical data, and a Shapiro-Wilk test with p > 0.05 will be used to define normal distribution. Univariate analysis will be conducted to test for statistically significant differences in observation scores between open radical cystectomy and robot-assisted radical cystectomy cohorts across all variables, using independent sample t tests and Mann-Whitney U testing, as appropriate. A variable-selection strategy will be used to create multivariate models. Binary logistic regression will be conducted to calculate odds ratios and 95% confidence intervals for significant predictors on univariate analysis and clinically relevant covariates. Statistical significance is set at p < 0.05 based on a two-tailed comparison. SUMMARY: This study uses a structured approach to the analysis of nontechnical skills using extracorporeal videos of both open radical cystectomy and robot-assisted radical cystectomy surgeries, in order to obtain detailed data on nontechnical skills during open and minimally invasive surgeries. The results of this study could possibly be used to develop team-training programmes, specifically for the introduction of the surgical robot in relation to changes in nontechnical skills. Additional analysis of technical skills using the intracorporeal footage of the surgical robot will be used to elucidate the role of surgical skills and surgical events in nontechnical skills.

Artificial intelligence and robotics: a combination that is changing the operating room.

PURPOSE: The aim of the current narrative review was to summarize the available evidence in the literature on artificial intelligence (AI) methods that have been applied during robotic surgery. METHODS: A narrative review of the literature was performed on MEDLINE/Pubmed and Scopus database on the topics of artificial intelligence, autonomous surgery, machine learning, robotic surgery, and surgical navigation, focusing on articles published between January 2015 and June 2019. All available evidences were analyzed and summarized herein after an interactive peer-review process of the panel. LITERATURE REVIEW: The preliminary results of the implementation of AI in clinical setting are encouraging. By providing a readout of the full telemetry and a sophisticated viewing console, robot-assisted surgery can be used to study and refine the application of AI in surgical practice. Machine learning approaches strengthen the feedback regarding surgical skills acquisition, efficiency of the surgical process, surgical guidance and prediction of postoperative outcomes. Tension-sensors on the robotic arms and the integration of augmented reality methods can help enhance the surgical experience and monitor organ movements. CONCLUSIONS: The use of AI in robotic surgery is expected to have a significant impact on future surgical training as well as enhance the surgical experience during a procedure. Both aim to realize precision surgery and thus to increase the quality of the surgical care. Implementation of AI in master-slave robotic surgery may allow for the careful, step-by-step consideration of autonomous robotic surgery.

High-precision Bladder Cancer Irradiation in the Elderly: Clinical Results for a Plan-of-the-day Integrated Boost Technique with Image Guidance Using Lipiodol Markers.

BACKGROUND: For most elderly patients with muscle-invasive bladder cancer (MIBC), surgery is not an option because of patient frailty. Conventional radiotherapy, with its high-dose irradiation of surrounding healthy tissues, remains the only curative treatment for this patient population. OBJECTIVE: To determine whether targeted radiotherapy with Lipiodol demarcation and plan-of-the-day integrated boost technique (LPOD) is a viable curative treatment for elderly patients with MIBC. DESIGN, SETTING, AND PARTICIPANTS: Between September 2008 and September 2016 all MIBC patients in our hospital were screened for eligibility. We included patients with localised, unifocal T2-T4N0M0 grade 2-3 MIBC. Patients with a tumour volume >50% of the bladder wall surface, previous pelvic radiotherapy, and unilateral or bilateral hip prostheses were excluded. INTERVENTION: Targeted radiotherapy using LPOD. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: Overall survival, urothelial cell cancer-specific survival (UCCSS), disease recurrence, and Radiation Therapy Oncology Group (RTOG) toxicity were measured. Statistical analyses included independent-sample t tests, χ2 tests, and Mann-Whitney U tests. RESULTS AND LIMITATIONS: A total of 44 patients (median age 80 yr) were included. Over median follow-up of 38 mo, one patient ceased treatment and 23 patients died. LPOD resulted in a 11.4% chance of local recurrence, high 3-yr UCCSS of 77%, RTOG grade >3 toxicity of 2.3-12.9%, and 3-yr overall survival of 49%. CONCLUSIONS: LPOD is a feasible first-line treatment option for older patients with limited-volume T2-T4N0M0 grade 2-3 MIBC. PATIENT SUMMARY: We looked at outcomes after targeted radiotherapy in elderly patients with muscle-invasive bladder cancer. We found that this treatment results in a low chance of disease recurrence with few toxicity complaints. We conclude that this treatment is a viable first-line treatment option for elderly patients.

The value of a 1-day multidisciplinary robot surgery training for novice robot surgeons.

INTRODUCTION: To fulfil the need for a basic level of competence in robotic surgery (Brinkman et al., Surg Endosc Other Interv Tech 31(1):281-287, 2017; Dutch Health inspectorate (Inspectie voor de gezondheidszorg), Insufficient carefulness at the introduction of surgical robots (in Dutch: Onvoldoende zorgvuldigheid bij introductie van operatierobots), Igz, Utrecht, 2010), the NIVEL (Netherlands Institute for Healthcare Research) developed the 'Basic proficiency requirements for the safe use of robotic surgery' (BPR). Based on the BPR a 1-day robotic surgery training was organised to answer the following research questions: (1) Are novice robot surgeons able to accurately self-assess their knowledge and dexterity skills? (2) Is it possible to include the teaching of all BPRs in a 1-day training? MATERIALS AND METHODS: Based on the BPR, a robot surgery course was developed for residents and specialists (surgery, gynaecology and urology). In preparation, the participants completed an online e-module. The 1-day training consisted of a practical part on robot set-up, a theoretical section, and hands-on exercises on virtual reality robot simulators. Multiple online questionnaire was filled out by the participants at the end of the training to evaluate the perceived educational value of the course and to self-assess the degree to which BPRs were reached. RESULTS: 20 participants completed the training during the conference of the Dutch Association for Endoscopic Surgery (NVEC) in 2017. Participants indicated nearly all competency requirements were mastered at the end of the training. The competency requirements not mastered were, however, critical requirements for the safe use of the surgical robot. Skill simulation results show a majority of participants are unable to reach a proficient simulation score in basic skill simulation exercises. CONCLUSION: Results show novice robot surgeons are too positive in the self-assessment of their own dexterity skills after a 1-day training. Self-assessment revealed uncertainty of the obtained knowledge level on requirements for the safe use of the surgical robot. Basic courses on robotic training should inform trainees about their results to enhance learning and inform them of their competence levels.