Description
Date depot: 1 janvier 1900
Titre: Integrative system for segmentation and tracking of 3D+t biological data
Directeur de thèse:
Thomas BOUDIER (IPAL (EDITE))
Domaine scientifique: Sciences et technologies de l'information et de la communication
Thématique CNRS : Non defini
Resumé:
Thanks to the development of new imaging techniques such as spinning-disk microscopy, confocal microscopy or Light Sheet
Fluorescence Microscopy (LSFM), acquiring data of live cells or organisms is more and more common in biological researches.
In order to obtain quantitativ results from 4D (3D+t) data, cells must be detected and tracked over time. Tracking relies on a prior segmentation or detection of the objects. However even state of the art segmentation algorithms may fail in detecting all cells correctly, especially in case of noisy data, or crowded environment. Usually only gap closings are correctly handled by tracking algorithms. Nevertheless other errors may occur like merged or splitted objects.
We want to propose a tracking procedure based on a frame by frame approach and post-processing of special events like merging and splitting. The tracking procedure should be able to detect errors in segmentation and give feedback to the segmentation algorithm to correct and improve the overall rate of correct segmentation and then tracking.
Some typical applications of cell tracking are the reconstruction of lineage in early embryogenesis, the study of cell dynamics or cell cycle (see for instance the cell tracking challenge in past ISBI conferences).
Doctorant.e: Nasser Khalafallah Mahmoud Lamees