RGB-D Camera Based Collision Prediction and Avoidance for X-ray Rotational Angiography
Kağan İncetan, Rishi Mohan, Henry Stoutjesdijk, Nelson Fernandes, Bram de Jager · 2019
Optimal clinical workflow leads to faster treatment times and a potential to cater to a larger number of patients. A key part of this is preventing collisions between moving medical systems and patient. For interventional environments, high-speed rotational angiography (RA) scans are prone to potential collisions between the C-arm X-ray system and the patient. To ensure safety, a low speed safety-run is executed before the actual high-speed scan. However, several iterations of the safety- run are often required before a scan is collision-free leading to a suboptimal clinical work-flow. This work proposes a RGB-D camera based collision prediction system which detects collisions before the actual RA scan. Additionally, a motion planner is designed to provide appropriate patient repositioning such that the collision is avoided. The system is introduced as a first proof- of-concept to eliminate the safety-run and improve clinical safety and workflow.