Related Experiment Video
Updated: Jan 7, 2026
10:25
Technical Approach for Infrared Tracking for Soft Tissue Navigation with a Holographic Head-Mounted Display and Preclinical Validation
Published on: September 2, 2025
449
A Deep Learning-Based Human-Robot Collaborative Navigation Framework for Vascular Interventional Surgery
Summary
This study introduces a deep learning framework for autonomous vascular interventional surgery (VIS). The system achieves 93.75% accuracy in catheter action decisions, improving surgical effectiveness and safety.
Area of Science:
- Robotics
- Artificial Intelligence
- Medical Technology
Background:
- Autonomous robotic surgery offers potential for improved patient outcomes.
- Vascular interventional surgery (VIS) presents unique challenges for autonomous execution due to dynamic surgical states and soft tissues.
Purpose of the Study:
- To develop an end-to-end deep learning framework for human-robot collaborative navigation in VIS.
- To enable autonomous execution of preplanned tasks in VIS.
Main Methods:
- A novel framework utilizing surgical Generative Adversarial Networks (GANs) for real-time catheter path planning.
- A Convolutional Neural Networks (CNNs)-based action estimator for nonlinear mapping.
- A human-robot trust-based shared control model for navigation.
Main Results:
- The system achieved a catheter action decision accuracy of 93.75% in experiments.
- Demonstrated improved surgical effectiveness and safety compared to existing methods.
- Networks were trained on a self-built dataset in a simulated catheterization environment.
Conclusions:
- The proposed framework offers a viable pathway towards achieving autonomous VIS.
- Highlights the potential of deep learning in overcoming challenges in complex surgical procedures.
