Global linking of cell tracks using the Viterbi algorithm.
Global linking of cell tracks using the Viterbi algorithm.
复制标题
DOI：
10.1109/tmi.2014.2370951
复制
发表时间：
2015-04
影响因子：
10.6
通讯作者：
Blau HM
中科院分区：
文献类型：
--
作者：
Magnusson KE;Jalden J;Gilbert PM;Blau HM
在显微镜图像序列中对活细胞的自动跟踪是一个重要且具有挑战性的问题。考虑到此应用程序，我们提出了一个链接算法的全局轨道，该算法将分割算法生成的细胞概述链接到轨道。该算法一次将轨道添加到图像序列一个中，以一种在每个链接决策中使用完整图像序列的信息。这是通过找到使用VITERBI算法的概率动机得分函数增加的轨道来实现的。我们还提出了一种新的方式，可以在创建新轨道时改变先前创建的轨道，从而减轻错误传播的影响。该算法可以处理有丝分裂，凋亡和迁移进出成像区域，还可以处理假阳性，遗漏的检测以及共同分段细胞的簇。在使用明亮场显微镜获得的两个具有挑战性的数据集中证明了算法性能，但原则上，该算法可以与任何单元格类型和任何成像技术一起使用，认为存在合适的分割算法。
Automated tracking of living cells in microscopy image sequences is an important and challenging problem. With this application in mind, we propose a global track linking algorithm, which links cell outlines generated by a segmentation algorithm into tracks. The algorithm adds tracks to the image sequence one at a time, in a way which uses information from the complete image sequence in every linking decision. This is achieved by finding the tracks which give the largest possible increases to a probabilistically motivated scoring function, using the Viterbi algorithm. We also present a novel way to alter previously created tracks when new tracks are created, thus mitigating the effects of error propagation. The algorithm can handle mitosis, apoptosis, and migration in and out of the imaged area, and can also deal with false positives, missed detections, and clusters of jointly segmented cells. The algorithm performance is demonstrated on two challenging datasets acquired using bright-field microscopy, but in principle, the algorithm can be used with any cell type and any imaging technique, presuming there is a suitable segmentation algorithm.