Achievement

A new approach has been developed that combines spatial information with a mixture of Gaussians model for automatic detection of live cell colonies and cells in videos of phase contrast images. This is an important problem for the quantification of HESCs in various states: attached, unattached, apoptosis and blebbing. Automatic cell detection is challenging because of low signal-to-noise ratio, the topological complexity of shapes and changes in intensity between cell and background over time. The method is validated using phase contrast images. It is compared with the current segmentation algorithms and found to produce the best results. A paper on this work has been accepted at the IEEE International Conference on Healthcare Informatics, Imaging and Systems Biology to be held in July 2011.