Guo Dongmin, van de Ven Anne L, Zhou Xiaobo
IEEE J Biomed Health Inform. 2014 May;18(3):991-8. doi: 10.1109/JBHI.2013.2281915. Epub 2013 Sep 16.
The investigation of microcirculation is an important task in biomedical and physiological research because the microcirculation information, such as flow velocity and vessel density, is critical to monitor human conditions and develop effective therapies of some diseases. As one of the tasks of the microcirculation study, red blood cell (RBC) tracking presents an effective approach to estimate some parameters in microcirculation. The common method for RBC tracking is based on spatiotemporal image analysis, which requires the image to have high qualification and cells should have fixed velocity. Besides, for in vivo cell tracking, cells may disappear in some frames, image series may have spatial and temporal distortions, and vessel distribution can be complex, which increase the difficulties of RBC tracking. In this paper, we propose an optical flow method to track RBCs. It attempts to describe the local motion for each visible point in the frames using a local displacement vector field. We utilize it to calculate the displacement of a cell in two adjacent frames. Additionally, another optical flow-based method, scale invariant feature transform (SIFT) flow, is also presented. The experimental results show that optical flow is quite robust to the case where the velocity of cell is unstable, while SIFT flow works well when there is a large displacement of the cell between two adjacent frames. Our proposed methods outperform other methods when doing in vivo cell tracking, which can be used to estimate the blood flow directly and help to evaluate other parameters in microcirculation.
微循环研究是生物医学和生理学研究中的一项重要任务，因为诸如流速和血管密度等微循环信息对于监测人体状况以及开发某些疾病的有效治疗方法至关重要。作为微循环研究的任务之一，红细胞（RBC）跟踪提供了一种估计微循环中某些参数的有效方法。红细胞跟踪的常用方法基于时空图像分析，这要求图像具有高质量，并且细胞应具有固定的速度。此外，对于体内细胞跟踪，细胞可能在某些帧中消失，图像序列可能存在时空扭曲，并且血管分布可能很复杂，这增加了红细胞跟踪的难度。在本文中，我们提出了一种光流方法来跟踪红细胞。它试图使用局部位移矢量场来描述帧中每个可见点的局部运动。我们利用它来计算细胞在两个相邻帧中的位移。此外，还提出了另一种基于光流的方法，尺度不变特征变换（SIFT）流。实验结果表明，光流对于细胞速度不稳定的情况非常鲁棒，而SIFT流在两个相邻帧之间细胞有大位移时效果良好。我们提出的方法在进行体内细胞跟踪时优于其他方法，可用于直接估计血流并有助于评估微循环中的其他参数。
