Realtime cell tracking in a microchannel

A method for simultaneously tracking multiple cells in a microchannel is proposed in this paper. The instantaneous size, position and velocity of cell though a microchannel can be obtained by using the method. In order to improve both the success rate and computing speed of the tracking, two possible fault cases are classified, and the fault cases are avoided during the image process. Red blood cells (RBCs) from the same subject on different days are tested in experiment. Result shows that the fault detection significantly improves the success rate of the tracking from 89.6% to 99.5%. The repeatability of the proposed method is verified by the similar distribution of cell size between the two tests. The proposed method is practically useful for evaluating cell characteristics in a microchannel system, and has potential toward a real-time cell tracking according to the low computing cost demonstrated in the experimental results.