SCH: Versatile and Compact Telerobot with Haptic Feedback and Physics-Informed Simulation for Safety Enhanced Neurovascular Interventions

Stroke is a leading cause of mortality and chronic disability, which predominately occurs due to blood clots or plaques impeding the blood flow in the brain (ischemic stroke) or a rupture in brain arteries (hemorrhagic stroke). Minimally invasive endovascular procedures, e.g., thrombectomy (removing clots to

restore blood flow) and neurovascular embolization (deploying coils in aneurysms to obstruct blood flow), are employed to treat these two strokes but also pose risks for X-ray radiation exposure. Robots were introduced to safeguard operators from radiation and increase precision via teleoperated control several

meters away from the patient. However, these robots are bulky capital equipment and only mechanically compatible with a few instruments. To tackle these limitations, we developed a versatile 4-DOF robot with a significantly smaller size than state-of-the-art robots and is compatible with a wide variety of

instruments. To further improve effectiveness and usability of endovascular robots, we identify three key limitations to the widespread adoption of robotic systems: 1) prolonged procedure time due to frequent robotic to manual conversions due to lack of full actuation of instruments; 2) steep learning curve and long

training time for clinicians to implement complex surgical manipulation with robots to perform interventions due to the lack of coordinated control and pre-operative training of robot-assisted procedure; 3) elevated risks of vessel or aneurysm ruptures due to lack of haptic feedback for instrument-vessel interaction force.

Our interdisciplinary team, consisting of experts in robotics (Su), mechanics (Jawed), interventional neuroradiology (Tateshima, M.D.), anatomy (Hartstone-Rose), human-robot interaction (Joo) will: 1) develop full actuation (avoid manual loading/unloading of instruments), coordinated control paradigms,

and patient-side haptic module for our robot; 2) establish a machine learning-assisted physics-based simulation framework for pre-operative training and intra-operative situational awareness; 3) study human-robot interaction to evaluate multiple metrics about manual, partial actuation, and full actuation

procedures. Our goal is to design intelligent robots in concert with pre-operative virtual training and intra-operative virtual fixtures (safety zone) to improve effectiveness and usability, thus ultimately enhancing safety and clinical outcome of neurovascular interventions. RELEVANCE (See instructions):

Our project entails transformative methods that span design, simulation, and medicine for an in-depth understanding of human-robot interaction for a surgery that saves the lives of millions of humans. Our work will advance two fundamental science disciplines: Robotics and Mechanics, and will make scientific

contributions to computer sciences and engineering to improve fundamental understanding of medicine.