A Versatile Centrifuge-Coupled Lab-on-a-Compact Disk Hardware for Quantification of Cell Volume Fractions by Electrical Impedance Spectroscopy

A versatile, centrifuge-coupled lab-on-a-compact disk (ccLoCD) hardware is developed for high through-put quantification of cell volume fractions ( $\phi $ ) by electrical impedance spectroscopy (EIS). The eight-electrode sensor hardware consists of 32-bit ARM microcontroller, impedance converter, and two high-speed analog multiplexer (MUX) modules on a disk mounted on a motor. It was evaluated by quantifying red blood cells (RBCs) of; 1. chicken ( $\phi _{\mathrm {chic}}^{\mathrm {RBC}}$ ) mixed with pork RBC with centrifugation and 2. pork RBC ( $\phi _{\mathrm {por}}^{\mathrm {RBC}}$ ) as they sediment into the sensor in saline (S), 50% plasma (50-P) and whole blood (WB) for assay of effects of protein concentration ( $\text {Conc}^{\mathrm {p}}$ ) on erythrocyte sedimentation rates (ESRs). To check quantification accuracy (QA), $\phi _{\mathrm {por}}^{\mathrm {RBC}}$ (blood hematocrit) in cardiopulmonary bypass (CPB) was measured in comparison to $\phi _{\mathrm {por}}^{\mathrm {RBC}}$ from standard microcentrifuge (MC) method at increasing plasma volumes ( $v_{\mathrm {p}}$ ). Blood impedance, measured between 10 and 100 kHz, was analyzed by distribution of relaxation times (DRTs) and regression modeling to quantify $\phi $ by correlating ( $R^{2}$ ) amplitude maxima ( $\gamma _{\max }$ ) to $\phi $ . From results at characteristic relaxation time $\tau _{\mathrm {c}} = 1\times 10^{-5}$ s, differences in $\gamma _{\max }$ before and after centrifugation were statistically significant ( $p \lt 0.05$ ) with centrifugation increasing QA thus raising $R^{2}$ from 0.0242 to 0.989. In 15 min, differences in $\phi _{\mathrm {por}}^{\mathrm {RBC}}$ for RBC sedimentation in S, 50%-P, and WB were also statistically significant ( $p \lt 0.05$ ) showing capacity of hardware to capture $\text {Conc}^{\mathrm {p}}$ effects. In QA analysis, hardware-based $\phi _{\mathrm {por}}^{\mathrm {RBC}}$ had low 0.73% error relative to MC-based $\phi _{\mathrm {por}}^{\mathrm {RBC}}$ . The $\phi _{\mathrm {por}}^{\mathrm {RBC}}$ strongly correlated with $v_{\mathrm {p}}$ in flow ( $R^{2} =0.9937$ ) which shows hardware’s potential for low cost measurements of cell $\phi $ .