|Title||Stem-cell localization: a deconvolution problem|
|Publication Type||Conference Paper|
|Year of Publication||2007|
|Authors||N Kachouie, N., P. Fieguth, and E. Jervis|
|Conference Name||29th Annual International Conference of the IEEE Engineering in Medicine and Biology Society|
|Keywords||adjacent cells, biomedical optical imaging, blood, cell segmentation, cellular biophysics, deconvolution, deconvolution problem, hematopoietic stem cells, image segmentation, immune cells, medical image processing, optimisation, optimization algorithm, stem-cell localization, template matching method|
Hematopoietic Stem Cells (HSCs) form blood and immune cells and are responsible for the constant renewal of blood. To produce new blood cells, HSCs proliferate and differentiate to different blood cell types continuously during their lifetime. Hence they are of substantial interest in stem cell therapy and cancer research. To classify HSCs to different groups, they must be observed/tracked over time and their key features including cell size, shape, and motility must be extracted. The manual tracking is an onerous task and automated methods are in high demand. The first stage of an semi-automatic/automatic tracking system is cell segmentation. In our previous work we addressed the cell segmentation/localization problem. Modelling adjacent or splitting cells is very challenging and our previous methods might fail to accurately model a group of adjacent cells or a splitting cell. In this paper we address this issue and propose a deconvolution method to precisely model individual HSCs as well as adjacent (splitting) HSCs. An optimization algorithm is combined with a template matching method to segment cell regions and locate the cell centers.
Stem-cell localization: a deconvolution problem