Cell Counting for in Vivo Flow Cytometry Signals with Baseline Drift

Xiaoling Wang,Yuanzhen Suo,Dan Wei,Hao He,Fan Wu,Xunbin Wei

DOI: https://doi.org/10.1142/s1793545817500080

2017-01-01

Abstract:In biomedical research fields, the in vivo flow cytometry (IVFC) is a widely used technology which is able to monitor target cells dynamically in living animals. Although the setup of IVFC system has been well established, baseline drift is still a challenge in the process of quantifying circulating cells. Previous methods, i.e., the dynamic peak picking method, counted cells by setting a static threshold without considering the baseline drift, leading to an inaccurate cell quantification. Here, we developed a method of cell counting for IVFC data with baseline drift by interpolation fitting, automatic segmentation and wavelet-based denoising. We demonstrated its performance for IVFC signals with three types of representative baseline drift. Compared with non-baseline-correction methods, this method showed a higher sensitivity and specificity, as well as a better result in the Pearson’s correlation coefficient and the mean-squared error (MSE).