Eindhoven University of Technology, Eindhoven, The Netherlands.
Philips Research, Eindhoven, The Netherlands.
Int J Comput Assist Radiol Surg. 2019 Jun;14(6):1069-1077. doi: 10.1007/s11548-019-01960-y. Epub 2019 Apr 9.
Efficient image-based catheter localization in 3D US during cardiac interventions is highly desired, since it facilitates the operation procedure, reduces the patient risk and improves the outcome. Current image-based catheter localization methods are not efficient or accurate enough for real clinical use.
We propose a catheter localization method for 3D cardiac ultrasound (US). The catheter candidate voxels are first pre-selected by the Frangi vesselness filter with adaptive thresholding, after which a triplanar-based ConvNet is applied to classify the remaining voxels as catheter or not. We propose a Share-ConvNet for 3D US, which reduces the computation complexity by sharing a single ConvNet for all orthogonal slices. To boost the performance of ConvNet, we also employ two-stage training with weighted cross-entropy. Using the classified voxels, the catheter is localized by a model fitting algorithm.
To validate our method, we have collected challenging ex vivo datasets. Extensive experiments show that the proposed method outperforms state-of-the-art methods and can localize the catheter with an average error of 2.1 mm in around 10 s per volume.
Our method can automatically localize the cardiac catheter in challenging 3D cardiac US images. The efficiency and accuracy localization of the proposed method are considered promising for catheter detection and localization during clinical interventions.
在心脏介入过程中，高效的基于图像的导管定位在 3D US 中是非常需要的，因为它可以简化操作流程，降低患者风险并改善结果。目前基于图像的导管定位方法在实际临床应用中效率或准确性还不够高。
我们提出了一种用于 3D 心脏超声（US）的导管定位方法。首先，通过自适应阈值的 Frangi 血管滤波器预选导管候选体素，然后应用基于三平面的 ConvNet 将其余体素分类为导管或非导管。我们提出了一种用于 3D US 的共享 ConvNet，通过为所有正交切片共享单个 ConvNet 来降低计算复杂度。为了提高 ConvNet 的性能，我们还采用了两阶段训练和加权交叉熵。使用分类的体素，通过模型拟合算法定位导管。
为了验证我们的方法，我们收集了具有挑战性的离体数据集。广泛的实验表明，该方法优于最先进的方法，并且可以在大约 10 秒内对每个体积进行平均误差为 2.1 毫米的导管定位。
我们的方法可以自动定位具有挑战性的 3D 心脏 US 图像中的心脏导管。所提出的方法的效率和准确性定位被认为在临床干预期间用于导管检测和定位是有前途的。
