Autonomous Catheterization with Open-source Simulator and Expert Trajectory
Primary Area: applications to robotics, autonomy, planning
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Keywords: Endovascular Navigation, Expert Trajectory
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2024/AuthorGuide.
TL;DR: Introducing the first open-source endovascular simulator and autonomous catheterization agent
Abstract: Endovascular robots have been actively developed in both academia and industry. However, progress toward autonomous catheterization is often hampered by the widespread use of closed-source simulators and physical phantoms. Additionally, the acquisition of large-scale data for machine learning algorithms training for endovascular robots is usually infeasible due to the expensive medical setup and complicated medical procedures. We introduce CathSim, the first open-source and real-time simulator for endovascular intervention to address these limitations. CathSim's emphasis on sample efficiency enables rapid development and testing of learning algorithms. We validate our CathSim with real robot experiments and achieve competitive performance. Based on CathSim, we develop a multi-modal expert navigation network and demonstrate its effectiveness in downstream endovascular navigation tasks. Our results suggest that CathSim has the potential to significantly accelerate research in the field of autonomous endovascular navigation systems. Our project is available at https://anonymous.4open.science/r/cathsim-E168.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors' identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 5177
Loading