A novel cell tracking algorithm and continuous hidden Markov model for cell phase identification

Time-lapse microscopy cell imaging is attracting more and more attentions due to its potential in achieving new and high throughput ways to conduct drug discovery and quantitative cellular studies. However, the lacking of effective automatic systems for studying a large population of cell nuclei is limiting the application of it. In this paper, we propose a novel hybrid merging algorithm for cell nuclei segmentation and propose a novel favorite matching plus local tree matching algorithm to track dynamic behaviors of a large population of cell nuclei in time-lapse microscopy. And then we propose to identify the phases of cell nuclei using context information of tracks by continuous hidden Markov model. Experimental results show the whole proposed system is very effective for time-lapse microscopy cell imaging segmentation, tracking and cell phase identification