Computer-aided analysis of cell interactions under dynamic flow conditions

Experimentally, initial steps of leucocyte extravasation, including tethering and rolling, are analysed in endothelial cell flow chambers. Given the complexity and speed of endothelial-immune cell interaction, computer-aided advances of this analysis are highly desirable. Herein, we compared two established methods, hand counting and tracking software, with novel analysis software using defined movies recorded at standard conditions of endothelial-leucocyte interactions. As a first validation, cell counts and velocity parameters determined by seven experienced experts revealed no statistic differences to both semi-automated tracking and fully computerized analyses. Nevertheless, interindividual variations were substantial for hand counting. In additional experiments, velocity distributions between 1 and 800 mu m/s picked up by the fully computerized analysis matched well with the tracking software as indicated by speed vector histograms. With respect to the time consumed for a defined set of movies, hand counting took 3.6 +/- 1.6 h, tracking software 4.5 +/- 1.2 h, whereas fully automated analysis consumed less than 15 min, reaching real-time mode. Thus, a validated and fully computerized method yielded functional flow chamber data unbiased, independent from an examiner, and reaching high-throughput level, which in turn will allow a substantial progress in understanding this process central for skin inflammation.