Artificial intelligence (AI) is revolutionizing nearly every aspect of modern life. In the medical field, robotic surgery is the sector with some of the most innovative and impactful advancements. In this narrative review, we outline recent contributions of AI to the field of robotic surgery with a particular focus on intraoperative enhancement. AI modeling is allowing surgeons to have advanced intraoperative metrics such as force and tactile measurements, enhanced detection of positive surgical margins, and even allowing for the complete automation of certain steps in surgical procedures. AI is also Query revolutionizing the field of surgical education. AI modeling applied to intraoperative surgical video feeds and instrument kinematics data is allowing for the generation of automated skills assessments. AI also shows promise for the generation and delivery of highly specialized intraoperative surgical feedback for training surgeons. Although the adoption and integration of AI show promise in robotic surgery, it raises important, complex ethical questions. Frameworks for thinking through ethical dilemmas raised by AI are outlined in this review. AI enhancements in robotic surgery is some of the most groundbreaking research happening today, and the studies outlined in this review represent some of the most exciting innovations in recent years.

This article describes a machine training algorithm for evaluating surgeon performance in robotic surgery that was developed using automated performance metrics

We evaluated the face, content and construct validity of the novel da Vinci® Skills Simulator™ using the da Vinci Si™ Surgeon Console as the surgeon interface. We evaluated a novel robotic surgical simulator for robotic surgery using the da Vinci Si Surgeon Console and Mimic™ virtual reality. Subjects were categorized as novice-no surgical training, intermediate-surgical training with fewer than 100 robotic cases or expert-100 or more primary surgeon robotic cases. Each participant completed 10 virtual reality exercises with 3 repetitions and a questionnaire with a 1 to 10 visual analog scale to assess simulator realism (face validity) and training usefulness (content validity). The simulator recorded performance based on specific metrics. The performance of experts, intermediates and novices was compared (construct validity) using the Kruskal-Wallis test. We studied 16 novices, 32 intermediates with a median surgical experience of 6 years (range 1 to 37) and a median of 0 robotic cases (range 0 to 50), and 15 experts with a median of 315 robotic cases (range 100 to 800). Participants rated the virtual reality and console experience as very realistic (median visual analog scale score 8/10) while expert surgeons rated the simulator as a very useful training tool for residents (10/10) and fellows (9/10). Experts outperformed intermediates and novices in almost all metrics (median overall score 88.3% vs 75.6% and 62.1%, respectively, between group p<0.001). We confirmed the face, content and construct validity of a novel robotic skill simulator that uses the da Vinci Si Surgeon Console. Although it is currently limited to basic skill training, this device is likely to influence robotic surgical training across specialties.

The intraoperative activity of a surgeon has substantial impact on postoperative outcomes. However, for most surgical procedures, the details of intraoperative surgical actions, which can vary widely, are not well understood. Here we report a machine learning system leveraging a vision transformer and supervised contrastive learning for the decoding of elements of intraoperative surgical activity from videos commonly collected during robotic surgeries. The system accurately identified surgical steps, actions performed by the surgeon, the quality of these actions and the relative contribution of individual video frames to the decoding of the actions. Through extensive testing on data from three different hospitals located in two different continents, we show that the system generalizes across videos, surgeons, hospitals and surgical procedures, and that it can provide information on surgical gestures and skills from unannotated videos. Decoding intraoperative activity via accurate machine learning systems could be used to provide surgeons with feedback on their operating skills, and may allow for the identification of optimal surgical behaviour and for the study of relationships between intraoperative factors and postoperative outcomes.

Objectives: To predict urinary continence recovery after robot‐assisted radical prostatectomy ( RARP ) using a deep learning ( DL ) model, which was then used to evaluate surgeon's historical patient outcomes. Subjects and Methods: Robotic surgical automated performance metrics ( APM s) during RARP , and patient clinicopathological and continence data were captured prospectively from 100 contemporary RARP s. We used a DL model (DeepSurv) to predict postoperative urinary continence. Model features were ranked based on their importance in prediction. We stratified eight surgeons based on the five top‐ranked features. The top four surgeons were categorized in ‘Group 1/ APM s’, while the remaining four were categorized in ‘Group 2/APMs’. A separate historical cohort of RARP s (January 2015 to August 2016) performed by these two surgeon groups was then used for comparison. Concordance index ( C ‐index) and mean absolute error ( MAE ) were used to measure the model's prediction performance. Outcomes of historical cases were compared using the Kruskal–Wallis, chi‐squared and Fisher's exact tests. Results: Continence was attained in 79 patients (79%) after a median of 126 days. The DL model achieved a C ‐index of 0.6 and an MAE of 85.9 in predicting continence. APM s were ranked higher by the model than clinicopathological features. In the historical cohort, patients in Group 1/ APM s had superior rates of urinary continence at 3 and 6 months postoperatively (47.5 vs 36.7%, P = 0.034, and 68.3 vs 59.2%, P = 0.047, respectively). Conclusion: Using APM s and clinicopathological data, the DeepSurv DL model was able to predict continence after RARP . In this feasibility study, surgeons with more efficient APM s achieved higher continence rates at 3 and 6 months after RARP .

We explore and validate objective surgeon performance metrics using a novel recorder ("dVLogger") to directly capture surgeon manipulations on the da Vinci® Surgical System. We present the initial construct and concurrent validation study of objective metrics during preselected steps of robot-assisted radical prostatectomy. Kinematic and events data were recorded for expert (100 or more cases) and novice (less than 100 cases) surgeons performing bladder mobilization, seminal vesicle dissection, anterior vesicourethral anastomosis and right pelvic lymphadenectomy. Expert/novice metrics were compared using mixed effect statistical modeling (construct validation). Expert reviewers blindly rated seminal vesicle dissection and anterior vesicourethral anastomosis using GEARS (Global Evaluative Assessment of Robotic Skills). Intraclass correlation measured inter-rater variability. Objective metrics were correlated to corresponding GEARS metrics using Spearman's test (concurrent validation). The performance of 10 experts (mean 810 cases, range 100 to 2,000) and 10 novices (mean 35 cases, range 5 to 80) was evaluated in 100 robot-assisted radical prostatectomy cases. For construct validation the experts completed operative steps faster (p <0.001) with less instrument travel distance (p <0.01), less aggregate instrument idle time (p <0.001), shorter camera path length (p <0.001) and more frequent camera movements (p <0.03). Experts had a greater ratio of dominant-to-nondominant instrument path distance for all steps (p <0.04) except anterior vesicourethral anastomosis. For concurrent validation the median experience of 3 expert reviewers was 300 cases (range 200 to 500). Intraclass correlation among reviewers was 0.6-0.7. For anterior vesicourethral anastomosis and seminal vesicle dissection, kinematic metrics had low associations with GEARS metrics. Objective metrics revealed experts to be more efficient and directed during preselected steps of robot-assisted radical prostatectomy. Objective metrics had limited associations to GEARS. These findings lay the foundation for developing standardized metrics for surgeon training and assessment.