An improved real-time endovascular guidewire position simulation using shortest path algorithm

Abstract

In this study, we propose a new graph-theoretical method to simulate guidewire paths inside the carotid artery. The minimum energy guidewire path can be obtained by applying the shortest path algorithm, such as Dijkstra's algorithm for graphs, based on the principle of the minimal total energy. Compared to previous results, experiments of three phantoms were validated, revealing that the first and second phantoms overlap completely between simulated and real guidewires. In addition, 95 % of the third phantom overlaps completely, and the remaining 5 % closely coincides. The results demonstrate that our method achieves 87 and 80 % improvements for the first and third phantoms under the same conditions, respectively. Furthermore, 91 % improvements were obtained for the second phantom under the condition with reduced graph construction complexity.

An improved matrix‐based endovascular guidewire position simulation using fusiform ternary tree

“…In this study, we present a new matrix-based method to simulate the guidewire behavior inside an arterial system. Compared with the previous method, 24 our method improves previous iterative refinement The number of simulation (Table 2) using TP, FN, FP, and OM show our method has a better local simulation accuracy than that of References 24 and 29 for three phantoms and eight Rotterdam datasets. Thus, the proposed method has a better robustness on the local performance for guidewire and its two ends.…”

“…In Reference 24, an iterative refinement approach was proposed but failed to accurately screen out some maximum curvature points. In this subsection, to remedy this disadvantage, we improve the approach by canceling the parameter α used in Reference 24 and by adopting an adaptive sampling strategy.…”

“…Besides, this model has been used in the surgical path planning of medical robots 22 and centerline extraction 23 . In Reference 24, an iterative refinement algorithm was proposed to further obtain an ideal running time according to the ideas and methods of Chen et al 25‐28 …”

“…In this study, we make contribution by proposing a novel matrix‐based method to simulate in real‐time the guidewire behavior inside an arterial system. Compared with the previous method, 24 the proposed method improves the iterative refinement procedure (ie, cancel the parameter α [the number of maximum curvature points] and adopt an adaptive sampling strategy), and adopts the fusiform ternary tree approach, as well as an optimization strategy based on the proximal end and distal end of the guidewire, by using a multi‐thread technique, to improve both running time and accuracy in simulations.…”

“…In this study, we make contribution by proposing a novel matrix-based method to simulate in real-time the guidewire behavior inside an arterial system. Compared with the previous method, 24…”