Integrated Robotic–Imaging Platforms in Endovascular Surgery: Current Capabilities and Future Directions

Abstract

Robotic–imaging integrated platforms have emerged as a transformative approach for enhancing precision, safety, and efficiency in endovascular interventions. The rapid evolution of catheter robotics, multimodal imaging fusion, and AI-assisted navigation has created new opportunities to overcome longstanding limitations associated with manual procedures, including operator variability, radiation exposure, and constrained visualization in complex vascular anatomies. This study develops and evaluates a modular robotic–imaging architecture incorporating fluoroscopy, cone-beam CT, IVUS, ultrasound, and sensor-based feedback to support real-time navigation and intraoperative decision-making. The methodology integrates system architecture design, sensing and signal processing, experimental trials using anatomically realistic vascular models, and AI-driven computational modeling for trajectory prediction and motion compensation. Quantitative metrics such as trajectory deviation, latency, force application, and sensor noise characterization were analyzed, while qualitative assessments captured operator usability and interface effectiveness. Experimental results demonstrate significant performance enhancements, including improved navigation accuracy approaching 1 mm, latency reduction to below 80 ms, and a 50% reduction in operator radiation exposure due to optimized imaging utilization and console-based control. Sensor fusion and adaptive control mechanisms contributed to 25–40% improvements in system stability and trajectory consistency. Haptic feedback integration, with force sensitivity reaching 0.1 N, enhanced procedural safety by reducing unintended vessel contact. AI-based navigation models further improved path prediction reliability under varying physiological conditions. Overall, the findings confirm that the proposed robotic–imaging platform offers substantial improvements in precision, safety, and workflow efficiency. Continued refinement in multimodal data fusion, real-time autonomy, and clinical workflow integration will be essential to advancing next-generation robotic endovascular systems toward widespread clinical translation.