Liang Liang, Shen Hongying, De Camilli Pietro, Toomre Derek K, Duncan James S
Yale University, New Haven, CT 06520, USA.
Med Image Comput Comput Assist Interv. 2011;14(Pt 1):629-36. doi: 10.1007/978-3-642-23623-5_79.
Multi-angle total internal reflection fluorescence microscopy (MA-TIRFM) is a new generation of TIRF microscopy to study cellular processes near dorsal cell membrane in 4 dimensions (3D+t). To perform quantitative analysis using MA-TIRFM, it is necessary to track subcellular particles in these processes. In this paper, we propose a method based on a MAP framework for automatic particle tracking and apply it to track clathrin coated pits (CCPs). The expectation maximization (EM) algorithm is employed to solve the MAP problem. To provide the initial estimations for the EM algorithm, we develop a forward filter based on the most probable trajectory (MPT) filter. Multiple linear models are used to model particle dynamics. For CCP tracking, we use two linear models to describe constrained Brownian motion and fluorophore variation according to CCP properties. The tracking method is evaluated on synthetic data and results show that it has high accuracy. The result on real data confirmed by human expert cell biologists is also presented.
多角度全内反射荧光显微镜（MA-TIRFM）是新一代的全内反射荧光显微镜，用于在四维（三维+时间）条件下研究背侧细胞膜附近的细胞过程。为了使用MA-TIRFM进行定量分析，有必要在这些过程中追踪亚细胞颗粒。在本文中，我们提出了一种基于MAP框架的自动颗粒追踪方法，并将其应用于追踪网格蛋白包被小窝（CCP）。期望最大化（EM）算法被用于解决MAP问题。为了给EM算法提供初始估计，我们基于最可能轨迹（MPT）滤波器开发了一种前向滤波器。多个线性模型被用于对颗粒动力学进行建模。对于CCP追踪，我们根据CCP的特性使用两个线性模型来描述受限布朗运动和荧光团变化。该追踪方法在合成数据上进行了评估，结果表明它具有很高的准确性。同时也展示了由细胞生物学专家确认的真实数据的结果。
