Toward Autonomous Cannulation in Endovascular Intervention

Recently, many robotic systems for endovascular intervention have been proposed
in academia or industry. These teleoperation platforms commonly
consist of master and slave robots. Hence, the device architectures address
low levels of robotic autonomy. Examples of novel robotic platforms are
Magellan and Sensei X2 (AurisHealth, USA), R-one (Robocath, France),
CorPath GRX (Corindus, USA), CathBot (Imperial College London, UK).
Clinical trials demonstrated the applicability of different platforms. Beyond
that, novel master interfaces were described for optimised teleoperation feedback,
transparency, and usability. Alternatively, manual manipulation of
standard catheters was sensed without feedback and was replicated to a
remote slave platform. Furthermore, designs of slave kinematics addressed
different electromechanical configurations for instrument manipulation, i.e.,
translation/rotation, and coupling interfaces. Within all those aspects, machine
learning has been used to improve the assistive features which include
deep learning applications to handle the tracking challenges implied by the
system. Those features allow the development of higher levels of autonomy.
This PhD project aims to develop autonomous cannulation and tracking solutions
based on machine learning, bioelectrical localization, and robotics.
The novel autonomous cannulation and tracking technology will be incorporated
within a state-of-the-art robotic steering system for cooperative
catheter insertion. The system will feature multiple degrees of freedom to
accurately control the insertion process, haptic feedback for an improved
user experience, and the option for manual or autonomous robotic insertion.
The result will be a novel, autonomous, and radiation-free system for
endovascular surgery.