Optical Flow-Based Tracking of Needles and Needle-Tip Localization Using Circular Hough Transform in Ultrasound Images.
Optical Flow-Based Tracking of Needles and Needle-Tip Localization Using Circular Hough Transform in Ultrasound Images.
复制标题
DOI：
10.1007/s10439-014-1208-0
复制
发表时间：
2015-08
影响因子：
3.8
通讯作者：
Desai, Jaydev P.
中科院分区：
文献类型：
--
作者：
Ayvali, Elif;Desai, Jaydev P.
图像引导的干预已经成为基于针的程序的护理标准。图像引导手术的成功取决于精确定位和跟踪针的能力。这项工作主要集中在二维超声为基础的跟踪空心针（套管），是由形状记忆合金致动器连接的直段。提出了一种基于光流场的面内跟踪算法，实现了对插管形状的实时跟踪。光流是一种强大的跟踪算法，可以很容易地在CPU上运行。然而，由于图像中的强度变化，该算法在直接应用于超声图像时表现不佳。在这项工作中提出的方法，使超声图像上使用的光流算法来跟踪针的功能。利用针尖倾斜的特点，采用圆Hough变换在离面成像时对针尖进行精确定位。通过组织模型内的实验和牛肾的离体实验，证明了所提出的跟踪方法的成功。使用在这项工作中提出的方法，定量信息的针配置是实时获得的，这是至关重要的生成控制输入的针和自动化的针插入。
Image-guided interventions have become the standard of care for needle-based procedures. The success of the image-guided procedures depends on the ability to precisely locate and track the needle. This work is primarily focused on 2D ultrasound-based tracking of a hollow needle (cannula) that is composed of straight segments connected by shape memory alloy actuators. An in-plane tracking algorithm based on optical flow was proposed to track the cannula configuration in real-time. Optical flow is a robust tracking algorithm that can easily run on a CPU. However, the algorithm does not perform well when it is applied to the ultrasound images directly due to the intensity variation in the images. The method presented in this work enables using the optical flow algorithm on ultrasound images to track features of the needle. By taking advantage of the bevel tip, Circular Hough transform was used to accurately locate the needle tip when the imaging is out-of-plane. Through experiments inside tissue phantom and ex-vivo experiments in bovine kidney, the success of the proposed tracking methods were demonstrated. Using the methods presented in this work, quantitative information about the needle configuration is obtained in real-time which is crucial for generating control inputs for the needle and automating the needle insertion.
登录
查看更多内容
影响因子：
10.9
作者：
Fichtinger, Gabor;Fiene, Jonathan P.;Kazanzides, Peter
通讯作者：
Kazanzides, Peter
影响因子：
10.9
作者：
Novotny, Paul M.;Stoll, Jeff A.;Howe, Robert D.
通讯作者：
Howe, Robert D.
影响因子：
8
作者：
Luengo Hendriks, CL;van Ginkel, M;van Vliet, LJ
通讯作者：
van Vliet, LJ
DOI：
10.1177/0278364912442429
发表时间：
2012-04
期刊：
The International journal of robotics research
影响因子：
--
作者：
Ayvali E;Liang CP;Ho M;Chen Y;Desai JP
通讯作者：
Desai JP
影响因子：
2.9
作者：
Golemati, Spyretta;Stoitsis, John;Nikita, Konstantina S.
通讯作者：
Nikita, Konstantina S.