IEEE Trans Med Imaging. 2019 Dec;38(12):2785-2795. doi: 10.1109/TMI.2019.2914074. Epub 2019 May 1.
Intravascular ultrasound (IVUS) is a widely used interventional imaging technique for the assessment of atherosclerosis plaque. Due to pulsatile heart motions, transverse and longitudinal motions are observed during in vivo pullbacks of IVUS sequences. These motion artifacts can mislead the volume-based data retrieved from IVUS studies and hinder the visualization of the vessel condition. To overcome this problem, a new fully automatic image-based gating algorithm was proposed in the current study. We utilized the phase information of the dual-tree complex wavelet transform (DT-CWT) coefficients to detect the motion of edge-like structures. For each IVUS sequence, first, six motion signals were detected by analyzing the phase of DT-CWT coefficients in six different directions. Then, the three best motion signals were selected by analyzing the frequency properties of each signal. Subsequently, these extracted signals were filtered using a modified Butterworth band-pass filter and the gated sequence was formed by using a combination of them. The proposed method was compared to four state-of-the-art methods and its frequency spectrum had more accurate characteristics in the cardiac frequency. In addition, the gated sequence extracted by the proposed method had the highest similarity to the extracted gated sequence by the physician.
血管内超声（IVUS）是一种广泛用于动脉粥样硬化斑块评估的介入成像技术。由于心脏的脉动运动，在 IVUS 序列的体内回拉过程中会观察到横向和纵向运动。这些运动伪影可能会误导从 IVUS 研究中检索到的基于体积的数据，并阻碍血管状况的可视化。为了解决这个问题，本研究提出了一种新的全自动基于图像的门控算法。我们利用双树复小波变换（DT-CWT）系数的相位信息来检测边缘状结构的运动。对于每个 IVUS 序列，首先通过分析 DT-CWT 系数在六个不同方向的相位来检测六个运动信号。然后，通过分析每个信号的频率特性来选择三个最佳的运动信号。然后，使用修改的巴特沃斯带通滤波器对这些提取的信号进行滤波，并使用它们的组合形成门控序列。将所提出的方法与四种最先进的方法进行了比较，其在心脏频率下的频谱具有更准确的特征。此外，所提出的方法提取的门控序列与医生提取的门控序列具有最高的相似度。
