Fig. 1.

Erysense device, principle of measurement, and AI–based analysis of single-cell microfluidic flow behavior of stored RBCs. (A) Image of the Erysense device. A micro-structured chip without the need to connect any tubes (i.e., no cleaning requirements, no risk for contamination) requires very small volumes of packed RBCs (1 µL). It is an all-integrated “point-of-care” device with a footprint of an Espresso machine. (B) Representative images of healthy croissant-shaped (Top Left) and slipper-shaped RBCs (Top Right) at low (1 mm/s) and high (10 mm/s) velocities, respectively. Representative images of pathological RBC shapes during capillary flow (Lower images). Flow is from Left to Right in all images. (C) AI-based evaluation of donor-dependent (donors 1 to 6) changes in the RBC microcapillary flow behavior. Dashed black lines correspond to linear fits. Gray areas indicate an estimate of a 95% prediction interval. The y-axis shows the logarithm of the fraction of pathological RBC shapes based on the shape phase diagrams. Panels A and B are reproduced with permission from Recktenwald et al. 2022 (4) and panel C with permission from Lopes et al. 2023 (6).