Fig. 11
From: Applications of machine learning for simulations of red blood cells in microfluidic devices
Changing the cell stiffness has a surprising effect on the trajectory differences. Due to the fact that the majority of cells in the channel passage utilise the sufficiently wide slit on the top of the channel, the average error value is very similar to the baseline comparison of simulations A50a and A50b. The differences in the simulations are reflected in the maximum error. Its rapid linear growth corresponds to the deviation of the predicted trajectories from the simulated trajectories of the stuck RBCs