A multiple hypothesis based method for particle tracking and its extension for cell segmentation.
A multiple hypothesis based method for particle tracking and its extension for cell segmentation.
复制标题
一种基于多个假设的粒子跟踪方法及其用于细胞分割的扩展。
DOI：
10.1007/978-3-642-38868-2_9
复制
发表时间：
2013
期刊：
影响因子：
--
通讯作者：
Duncan, James S
中科院分区：
文献类型：
--
作者：
Liang, Liang;Shen, Hongying;Rompolas, Panteleimon;Greco, Valentina;De Camilli, Pietro;Duncan, James S
在生物学研究中，经常需要跟踪显微图像中的数千个小颗粒，以分析细胞和亚细胞过程的潜在机制，这可能会导致更好地理解某些疾病过程。在本文中，我们提出了一种自动粒子跟踪方法，并将其应用于分析一个重要的亚细胞过程，即网格蛋白介导的内吞作用，使用全内反射显微镜。粒子检测通过使用图像滤波器和随后的高斯混合模型拟合，以实现亚像素分辨率。设计了一个基于多假设的框架来解决数据关联问题和处理拆分/合并事件。在不同场景下的合成数据上演示了跟踪方法，并将其应用于真实的数据。我们还表明，通过配备一个细胞检测模块，该方法可以直接扩展到分割细胞图像的双光子激发显微镜。
In biological studies, it is often required to track thousands of small particles in microscopic images to analyze underlying mechanisms of cellular and subcellular processes which may lead to better understanding of some disease processes. In this paper, we present an automatic particle tracking method and apply it for analyzing an essential subcellular process, namely clathrin mediated endocytosis using total internal reflection microscopy. Particles are detected by using image filters and subsequently Gaussian mixture models are fitted to achieve sub-pixel resolution. A multiple hypothesis based framework is designed to solve data association problems and handle splitting/merging events. The tracking method is demonstrated on synthetic data under different scenarios and applied to real data. We also show that, by equipping with a cell detection module, the method can be extended straightforwardly for segmenting cell images taken by two-photon excitation microscopy.