A Multi-Stage Guidewire Tip Tracking Framework for Cardiovascular Robotic Interventions

Jianjun Zhu,Cheng Wang,Guifang Zhang,Sitong Anna Teng,Ligong Lu,Gaojun Teng

DOI: https://doi.org/10.1109/cisp-bmei53629.2021.9624418

2021-01-01

Abstract:Guide-wire segmentation enhances its visibility for interventionalist and provides a safety-critical mechanism for robot-assisted interventions. However, it is challenging to distinguish guide-wire tips from other curvilinear structures in the field of view. This paper proposed a multi-stage framework for guide-wire tip tracking. The approach proceeds into three stages, 1) Prediction: detecting all potential guidewire-like feature clusters from the first frame; 2) Locking: eliminating non-guidewire features; 3) Tracking: tracking real guide-wire tip identified after the Locking stage and monitoring if unfeasible tortuous bending occurs during the intervention. Functions of guide-wire tip fold-back bending detection and bending energy calculation are designed for robotic safety-critical mechanisms. Our method is validated on multi-centre cardiovascular datasets from three hospitals. On average, it achieves 0.915 in F1-Score, 2.14 pixels in Fréchet distance, and 91.9% tracking accuracy, which refers to polygon-mask inclusion defined in this work. The motion-based method can eliminate all non-guidewire features earlier (for simple cases) or later but more accurately (for challenging cases). Based on the tracking result, fold-back bending events can be accurately detected. The average elapsed time of the first two stages is 116 ms, and one Tracking step costs 76 ms. The accuracy and computational time of the guide-wire tip tracking framework satisfy the cardiovascular robotic system's requirements.