Directly to content
  1. Publishing |
  2. Search |
  3. Browse |
  4. Recent items rss |
  5. Open Access |
  6. Jur. Issues |
  7. DeutschClear Cookie - decide language by browser settings

Probabilistic Tracking and Behavior Identification of Fluorescent Particles

Godinez, William J.

PDF, English - main document
Download (6MB) | Terms of use

Citation of documents: Please do not cite the URL that is displayed in your browser location input, instead use the persistent URL or the URN below, as we can guarantee their long-time accessibility.


Explicit and tractable characterizations of the dynamical behavior of virus particles are pivotal for a thorough understanding of the infection mechanisms of viruses. This thesis deals with the problem of extracting symbolic representations of the dynamical behavior of fluorescent particles from fluorescence microscopy image sequences. The focus is on the behavior of virus particles such as fusion with the cell membrane. A numerical representation is obtained by tracking the particles in the image sequences. We have investigated probabilistic tracking approaches, including approaches based on the Kalman filter as well as based on particle filters. For reasons of efficiency and robustness, we developed a tracking approach based on the probabilistic data association (PDA) algorithm in combination with an ellipsoidal sampling scheme that exploits effectively the image data via parametric appearance models. To track objects in close proximity, we compute the support that each image position provides to each tracked object relative to the support provided to the object's neighbors. After tracking, the problem of mapping the trajectory information computed by the tracking approaches to symbolic representations of the behavior arises. To compute symbolic representations of behaviors related to the fusion of single virus particles with the cell membrane based on their intensity over time, we developed a layered probabilistic approach based on stochastic hybrid systems as well as hidden Markov models (HMMs). We use a maxbelief strategy to efficiently combine both representations. The layered approach describes the intensity, intensity models, and behaviors of single virus particles. We introduce models for the evolution of the intensity and the behavior. To compute estimates for the intensity, intensity models, and behaviors we use a hybrid particle filter and the Viterbi algorithm. The developed approaches have been applied to synthetic images as well as to real microscopy image sequences displaying human immunodeficiency virus (HIV-1) particles. We have performed an extensive quantitative evaluation of the performance and a comparison with several existing approaches. It turned out that our approaches outperform previous ones, thus yielding more accurate and more reliable information about the behavior of virus particles. Moreover, we have successfully applied our tracking approaches to 3D image sequences displaying herpes simplex virus (HSV) replication compartments. We also applied the tracking approaches to image data displaying microtubule tips and analyzed their motion. In addition, our tracking approaches were successfully applied to the 2D and 3D image data of a Particle Tracking Challenge.

Item Type: Dissertation
Supervisor: Rohr, PD Dr. Karl
Date of thesis defense: 17 July 2013
Date Deposited: 09 Aug 2013 06:53
Date: 16 April 2013
Faculties / Institutes: The Faculty of Mathematics and Computer Science > Department of Computer Science
Subjects: 004 Data processing Computer science
Controlled Keywords: Bildverarbeitung, Objektverfolgung, Viren, Mikroskopie
Uncontrolled Keywords: Biomedical imaging, microscopy images, tracking, virus particles, behavior identification.
About | FAQ | Contact | Imprint |
OA-LogoDINI certificate 2013Logo der Open-Archives-Initiative