Cardiovascular disease is one of the primary causes of the death in the developed world. A significant subset of these diseases is treated with endovascular surgery using a catheter and guidewire combination. The manual insertion of surgical instruments into the veins of patients carries risks towards the structural integrity of the blood vessels, and therefore the health of the patient.
This thesis describes a hardware-in-the-loop framework including a robotic control platform that mimicks real surgical instrument handling in the SOFA simulation environment, connected and controlled via Python. The developed framework includes instrument force assessment, bezier spline-based pathing for instruments and a test ground with the aim to prevent dangerous situations from occurring via a combined approach of haptic feedback, input-blending shared control and visual guidance. Furthermore, the thesis presents the base for a surgical training network in simulation that can offer real-time visual and haptic feedback during the simulated procedure. Framework and control performance were evaluated qualitatively and quantitatively on different operators (N=5) during two separate trials with three and four different levels of feedback and guidance, respectively.
From the second set of trials is concluded that the implemented control method serves to reduce the amount of navigational errors, and that haptic feedback and danger-weighted shared control can prevent tip forces from escalating. During qualitative evaluation of framework performance, it is concluded that the core framework performs as intended and forms a solid base to expand upon for both control development and surgical training purposes.