Technology Category
Intelligent Robots: State EstimationComputer Vision: Biometrics, Face, Gesture & PoseMachine Learning: Calibration & Uncertainty Quantification
Pose State Perception of Interventional Robot for Cardio-cerebrovascular Procedures
🤖 AI Summary
To address the clinical limitations of sensor-dependent robot pose estimation in cardiovascular and cerebrovascular interventional procedures—namely, insufficient real-time performance and positional accuracy—this paper proposes a markerless, vision-only end-to-end pose estimation method. Our approach introduces a novel tri-module collaborative framework: (1) a dual-head multi-task U-Net for joint segmentation of vasculature and interventional devices (catheters/guidewires); (2) a topology-preserving skeletonization algorithm to enhance structural robustness; and (3) a lightweight pose regression network leveraging geometric feature modeling and analytical visual pose solving. Evaluated on real interventional fluoroscopic sequences, the method achieves sub-millimeter localization accuracy (<0.3 mm) and angular precision (<1.2°), while sustaining real-time tracking at ≥30 fps. This significantly improves navigation reliability and clinical practicality under sensor-free conditions.
Technology Category
Application Category
📝 Abstract
In response to the increasing demand for cardiocerebrovascular interventional surgeries, precise control of interventional robots has become increasingly important. Within these complex vascular scenarios, the accurate and reliable perception of the pose state for interventional robots is particularly crucial. This paper presents a novel vision-based approach without the need of additional sensors or markers. The core of this paper's method consists of a three-part framework: firstly, a dual-head multitask U-Net model for simultaneous vessel segment and interventional robot detection; secondly, an advanced algorithm for skeleton extraction and optimization; and finally, a comprehensive pose state perception system based on geometric features is implemented to accurately identify the robot's pose state and provide strategies for subsequent control. The experimental results demonstrate the proposed method's high reliability and accuracy in trajectory tracking and pose state perception.
Problem
Research questions and friction points this paper is trying to address.
Accurate pose perception for vascular interventional robots
Vision-based method without extra sensors or markers
Reliable robot control in complex vascular environments
Innovation
Methods, ideas, or system contributions that make the work stand out.
Vision-based approach without additional sensors
Dual-head multitask U-Net for vessel and robot detection
Geometric feature-based pose state perception system
🔎 Similar Papers
No similar papers found.
