|Quantitative Image Analysis - CNRS URA 2582|
|HEAD||Dr OLIVO-MARIN Jean-Christophe / email@example.com|
|MEMBERS||BERLEMONT Sylvain / CHENOUARD Nicolas / Dr MEAS-YEDID Vannary
Dr de CHAUMONT Fabrice / ZHANG Bo / Dr ZIMMER Christophe
The scientific project of the Quantitative Image Analysis (QuIA) unit is to develop image analysis and computer vision tools for the processing and quantification of multi-channel temporal 3D sequences in biological microscopy. Our work over the last years has been centered about developing new algorithms for multi-particle tracking, active contours models, PSF approximations for image deconvolution and image segmentation. It has resulted in powerful tools for spot detection and counting in real-time imaging of virus and genes, movement and shape analysis in 3D+t microscopy and cell growth analysis. These methods and algorithms have been instrumental for the successful achievement of a large number of biological projects to which we collaborated.
Multiple spot tracking in 3D+t microscopy
We have developed a method to perform the detection and the tracking of microscopic spots directly on four dimensional (3D+t) image data. It extends our previous work by being able to detect with high accuracy multiple biological objects moving in three-dimensional space and by incorporating the possibility to follow moving spots switching between different dynamics charac-teristics. Our method is based on a two step procedure: first, the objects are detected in the image stacks thanks to a procedure based on a three-dimensional wavelet transform; then the tracking is performed within a Bayesian framework where each object is represented by a state vector evolving according to biologically realistic dynamic models. The main advantage of wavelet-based detection is to be robust to the local variation of contrast and to the imaging noise. The Bayesian tracking allows predicting the new position of a spot knowing its past positions and increases the reliability of the data association step.
Cell shape and motility analysis
The methodological basis for our work on cell shape and motility is the framework of deformable models or active contours. A major progress made over the last years is the extension of our techniques to 3D+t data. We have taken advantage of the level set method and its relatively easy extension to higher dimensions to include most of the features pre-viously developed for 2D images, including a robust region model and multiple level sets coupled by a non-overlap constraint. In addition, the 3D method takes advantage of cell volume homeostasis to substantially improve the outlining of touching cells. This is an important step towards quantifying the behaviour of cells imaged in 3D environments, e.g. parasites in host tissue.
|Publications 2006 of the unit on Pasteur's references database|
Activity Reports 2006 - Institut Pasteur
If you have problems with this Web page, please write to firstname.lastname@example.org