Wen Quan, Gao Jean
Computer Science and Engineering, University of Electronic Science and Technology of China, Chengdu, Sichuan, China.
Annu Int Conf IEEE Eng Med Biol Soc. 2009;2009:1465-8. doi: 10.1109/IEMBS.2009.5332435.
With the introduction of sensitive and fast electronic imaging devices and the development of biological methods to tag proteins of interest by green fluorescent proteins (GFP), it has now become critical to develop automatic quantitative data analysis tools to study the live cell dynamics at subcellular level. In this paper, a sequential Monte Carlo (SMC) method to track variable number of multiple 3D subcellular structures is proposed. First, multiple subcellular structures are represented by a joint state. Then the distribution of the dimension changing joint state is sampled efficiently by the reverse jump Markov chain Monte Carlo (RJMCMC) method designed with update move, identity switch move, disappearing move, and appearing move. The experimental results show that the proposed method can successfully track multiple 3D subcellular structures with different motion modalities such as object appearing and disappearing.
随着灵敏快速的电子成像设备的引入以及利用绿色荧光蛋白（GFP）标记感兴趣蛋白质的生物学方法的发展，开发自动定量数据分析工具以研究亚细胞水平的活细胞动力学变得至关重要。本文提出了一种用于跟踪可变数量的多个三维亚细胞结构的序贯蒙特卡罗（SMC）方法。首先，多个亚细胞结构由一个联合状态表示。然后，通过设计有更新移动、恒等切换移动、消失移动和出现移动的可逆跳跃马尔可夫链蒙特卡罗（RJMCMC）方法，有效地对维度变化的联合状态的分布进行采样。实验结果表明，所提出的方法能够成功跟踪具有不同运动模式（如物体出现和消失）的多个三维亚细胞结构。
