Study on method of tracking the active cells in image sequences based on EKF-PF / 生物医学工程学杂志

In cell image sequences, due to the nonlinear and nonGaussian motion characteristics of active cells, the accurate prediction and tracking is still an unsolved problem. We applied extended Kalman particle filter (EKF-PF) here in our study, attempting to solve the problem. Firstly we confirmed the existence and positions of the active cells. Then we established a motion model and improved it via adding motion angle estimation. Next we predicted motion parameters, such as displacement, velocity, accelerated velocity and motion angle, in region centers of the cells being tracked. Finally we obtained the motion traces of active cells. There were fourteen active cells in three image sequences which have been tracked. The errors were less than 2.5 pixels when the prediction values were compared with actual values. It showed that the presented algorithm may basically reach the solution of accurate predition and tracking of the active cells.

Tracking of neural stem cells in high density image sequence based on Topological constraint combined with Hungarian algorithm / 生物医学工程学杂志

Analysis of neural stem cells' movements is one of the important parts in the fields of cellular and biological research. The main difficulty existing in cells' movement study is whether the cells tracking system can simultaneously track and analyze thousands of neural stem cells (NSCs) automatically. We present a novel cells' tracking algorithm which is based on segmentation and data association in this paper, aiming to improve the tracking accuracy further in high density NSCs' image. Firstly, we adopted different methods of segmentation base on the characteristics of the two cell image sequences in our experiment. Then we formed a data association and constituted a coefficient matrix by all cells between two adjacent frames according to topological constraints. Finally we applied The Hungarian algorithm to implement inter-cells matching optimally. Cells' tracking can be achieved according to this model from the second frame to the last one in a sequence. Experimental results showed that this approaching method has higher accuracy compared with that using the topological constraints tracking alone. The final tracking accuracies of average of sequence I and sequence II have been improved 10.17% and 4%, respectively.

Precise segmentation of cleavage of the neuron stem cells in time lapse image sequences / 生物医学工程学杂志

To study the cleavage of neuron stem cells in time lapse image sequences and realize their features abstraction, identification and tracking, a precise segmentation algorithm that can preserve the shape of division cells is presented in this paper. The fuzzy threshold segmentation is based on Zadth's maximum entropy. The optimal parameters of the maximum fuzzy entropy are decided by genetic algorithm. Region merging and splitting of the under-segmentation objects of the result of fuzzy segmentation are realized by weighted distance transform, region labeling and some operations on morphology. By comparison with some results of fuzzy and hard segmentation, this algorithm can implement the precise segmentation that is necessary for some specified objects in automatic identification and tracking of neuron stem cells.