Novel System for Vision-Based Tracking of Surgical Motion During Live Open-Heart Surgery

While current assessments of surgeons rely on human observations, controlled laboratory settings, or tracking technologies that aren’t usable in live operating theaters, motion-tracking systems have the potential to “assess intra-operative performance, provide feedback, and predict outcomes” during live surgery, according to Maj Stenmark, PhD, from the Children’s Heart Center of Skåne University Hospital in Lund, Sweden. In a report, published in The Journal of Surgical Research, Dr. Stenmark and colleagues described their experience developing a novel technique for surgical assessment using motion tracking. They advanced that their system was feasible for use during live open-heart surgery and stated that motion data was collected with “acceptable accuracy and speed at low computational cost.”

To facilitate tracking, Dr. Stenmark and collaborators designed a three-dimensional (3D)-printed “tracking die” to be fitted to DeBakey forceps during surgery. The operating table was recorded from above, and the die was tracked with software developed by the researchers. The system was tested in five open-heart procedures where surgeons were asked to report their subjective concerns and describe the weight of the die during the procedure. The accuracy of the system was compared against a robot-generated ground truth.

The article relayed that the 3D-printed tracking die weighed 6 grams, which increased the mass of the forceps by approximately 13%. According to the report, surgeons reported sensing a shift in the balance of the instrument, but in blind testing they were unable to correctly verify changes in weight. Regarding tracking accuracy, when two or more markers on the die were detected, the 3D position estimate was within 2 to 3 mm, on average, and within 1.1 to 2.6 degrees from the ground truth. Last, the computational time was 30 to 50 ms per frame on a standard laptop computer.

Overall, the study’s collaborators advanced that their tracking system was applicable for live surgeries “with negligible inconvenience to the surgeons.” They concluded that, “if validated, vision-based motion tracking of live surgery may contribute to an objective assessment of surgeons’ technical performance during specific surgical procedures and be used for monitoring of progress over time.”