Department of Mechanical and Aerospace Engineering, State University of New York at Buffalo, Buffalo, NY 14214, USA.
Int J Comput Assist Radiol Surg. 2009 Nov;4(6):597-608. doi: 10.1007/s11548-009-0385-z. Epub 2009 Jul 18.
Treatment of vascular disease often involves endovascular interventions which use the vascular system for delivering treatment devices via a previously inserted guidewire to the diseased site. Previous studies show relative reproducibility of guidewire position after insertion, indicating that the guidewire position is constrained and could be represented by an energy minimization approach. Such representation would support the surgeon's decision process in guidewire selection.
In this paper, we determine the guidewire position using a k-level graph based on 3D vessel information. Guidewire properties are incorporated into the graph as edge weights given by the local bending energy related to the local bending angle. The optimal path through this weighted directed graph is determined using a shortest path algorithm. Volumetric data of two different internal carotid artery phantoms (Ø 3.5-4.6 mm) was acquired. Two guidewires (Ø 0.33 mm) of different material properties (stainless steel, plastic-coated steel core) were inserted into the phantoms.
The average RMS distance between actual and simulated guidewire positions varies from 0.9 mm (plastic coated) to 1.3 mm (stainless steel); the computation time to determine the position was <2s.
The results indicate that the proposed technique yields reproducible and accurate guidewire positions within a short, clinically relevant time frame. These calculated positions may be useful in facilitating neurovascular interventions.
血管疾病的治疗通常涉及血管内介入，通过先前插入的导丝将治疗设备输送到病变部位。先前的研究表明导丝插入后的位置具有相对可重复性，这表明导丝位置受到约束，可以通过能量最小化方法来表示。这种表示方法将支持外科医生在导丝选择方面的决策过程。
在本文中，我们使用基于 3D 血管信息的 k-级图来确定导丝的位置。将导丝特性作为与局部弯曲角相关的局部弯曲能的边缘权重纳入到图中。通过最短路径算法确定通过这个加权有向图的最优路径。获取了两个不同的颈内动脉模型（Ø 3.5-4.6mm）的容积数据。将两根不同材料特性的导丝（Ø 0.33mm）插入到模型中。
实际和模拟导丝位置之间的平均 RMS 距离从 0.9mm（涂塑）到 1.3mm（不锈钢）不等；确定位置的计算时间<2s。
结果表明，所提出的技术可在短时间内（临床相关时间范围内）生成可重复且准确的导丝位置。这些计算出的位置可能有助于神经血管介入。
