A learning robot for cognitive camera control in minimally invasive surgery

Background We demonstrate the first self-learning, context-sensitive, autonomous camera-guiding robot applicable to minimally invasive surgery. The majority of surgical robots nowadays are telemanipulators without autonomous capabilities. Autonomous systems have been developed for laparoscopic camera guidance, however following simple rules and not adapting their behavior to specific tasks, procedures, or surgeons. Methods The herein presented methodology allows different robot kinematics to perceive their environment, interpret it according to a knowledge base and perform context-aware actions. For training, twenty operations were conducted with human camera guidance by a single surgeon. Subsequently, we experimentally evaluated the cognitive robotic camera control. A VIKY EP system and a KUKA LWR 4 robot were trained on data from manual camera guidance after completion of the surgeon’s learning curve. Second, only data from VIKY EP were used to train the LWR and finally data from training with the LWR were used to re-train the LWR. Results The duration of each operation decreased with the robot’s increasing experience from 1704 s ± 244 s to 1406 s ± 112 s, and 1197 s. Camera guidance quality (good/neutral/poor) improved from 38.6/53.4/7.9 to 49.4/46.3/4.1% and 56.2/41.0/2.8%. Conclusions The cognitive camera robot improved its performance with experience, laying the foundation for a new generation of cognitive surgical robots that adapt to a surgeon’s needs.

Robotic infant surgery with 3 mm instruments: a study in piglets of less than 10 kg body weight

No data exist concerning the appication of a new robotic system with 3 mm instruments (Senhance®, Transenterix) in infants and small children. Therefore, the aim of this study was to test the system for its feasibility, performance and safety of robotic pediatric abdominal and thoracic surgery in piglets simulating infants with a body weight lower than 10 kg. 34 procedures (from explorative laparoscopy to thoracoscopic esophageal repair) were performed in 12 piglets with a median age of 23 (interquartile range: 12–28) days and a median body weight of 6.9 (6.1–7.3) kg. The Senhance® robotic system was used with 3 mm instruments, a 10 mm 3D 0° or 30° videoscope and advanced energy devices, the setup consisted of the master console and three separate arms. The amount, size, and position of the applied ports, their distance as well as the distance between the three operator arms of the robot, external and internal collisions, and complications of the procedures were recorded and analyzed. We were able to perform all planned surgical procedures with 3 mm robotic instruments in piglets with a median body weight of less than 7 kg. We encountered two non-robot associated complications (bleeding from the inferior caval and hepatic vein) which led to termination of the live procedures. Technical limitations were the reaction time and speed of robotic camera movement with eye tracking, the excessive bending of the 3 mm instruments and intermittent need of re-calibration of the fulcrum point. Robotic newborn and infant surgery appears technically feasible with the Senhance® system. Software adjustments for camera movement and sensitivity of the fulcrum point calibration algorithm to adjust for the increased compliance of the abdominal wall of infants, therefore reducing the bending of the instruments, need to be implemented by the manufacturer as a result of our study. To further evaluate the Senhance® system, prospective trials comparing it to open, laparoscopic and other robotic systems are needed.

A Robotic Camera Holder Controlled by Head Movements: Exploring This New Robot-Surgeon Interface

Background. Most robotic camera steering devices (RCSDs) require active steering by the surgeon and necessarily increase workload. Clinical experience shows that standard laparoscopic procedures can be performed safely as solo surgery aided by RCSDs. No evidence exists concerning exploratory or emergency procedures. We compared the performance during unexpected laparoscopic tasks on surgical simulators aided either by an RCSD controllable by head movements of the surgeon or by a human camera assistant. Methods. Forty-five medical students without previous experience with minimal invasive surgery were randomized in 2 groups, and they performed standard and unexpected laparoscopic tasks requiring complex camera movements on box trainers either using an RCSD or assisted by a human camera assistant. Efficiency and performance parameters were recorded. Results. Performance in simulated standard procedures was equivalent. In simulated exploratory procedures, we saw significantly better performance scores in the conventional group versus the RCSD group. The strongest factor for these differences was the longer camera-adjusting time in the RCSD group versus the conventional group (PEG task = 208 ± 51 seconds vs 170 ± 36 seconds, P = .005; suture task = 563 ± 126 seconds vs 454 ± 201 seconds, P = .041). Conclusion. These results, obtained on surgical simulators, indicate that the solo approach to standard surgical tasks, facilitated by an RCSD controllable by head movements, can most likely be viewed as safe. Exploratory procedures with a relevant chance for complications or procedures that require rapid, often, or complex camera movements should rather be performed with a human camera assistant.