Robotic Endoscope Control Via Autonomous Instrument Tracking

Endoscope Visual Servoing
DOI: 10.3389/frobt.2022.832208 Publication Date: 2022-04-11T04:19:38Z
ABSTRACT
Many keyhole interventions rely on bi-manual handling of surgical instruments, forcing the main surgeon to a second act as camera assistant. In addition burden excessively involving staff, this may lead reduced image stability, increased task completion time and sometimes errors due monotony task. Robotic endoscope holders, controlled by set basic instructions, have been proposed an alternative, but their unnatural increase cognitive load (solo) surgeon, which hinders clinical acceptance. More seamless integration in workflow would be achieved if robotic holders collaborated with operating via semantically rich instructions that closely resemble otherwise issued human assistant, such "focus my right-hand instrument". As proof concept, paper presents novel system paves way towards synergistic interaction between surgeons holders. The platform allows perform bimanual coordination navigation task, while arm autonomously performs positioning tasks. Within our system, we propose tooltip localization method based tool segmentation visual servoing approach ensures smooth appropriate motion camera. We validate vision pipeline run user study system. relevance is ensured through use laparoscopic exercise validated European Academy Gynaecological Surgery involves navigation. Successful application provides promising starting point broader adoption
SUPPLEMENTAL MATERIAL
Coming soon ....
REFERENCES (86)
CITATIONS (28)
EXTERNAL LINKS
PlumX Metrics
RECOMMENDATIONS
FAIR ASSESSMENT
Coming soon ....
JUPYTER LAB
Coming soon ....