Figure 3
Development of eQUANT algorithms and evaluating performance. (A) Schematic of the look-up table generation process for a single strain, starting with bacterial colonies from a blood agar plate. Briefly, (1) colonies from the plate are suspended in growth medium at a concentration of ~ 1 × 107 CFU/mL. (2) The ORP sensor is introduced to the sample which is incubated at 37 °C. (3) During growth, both ORP signal (blue) and growth (red) are measured. (4) Growth and ORP data are collected for many strains of a bacterial species. These results are combined in order to associate ORP signal (ΔORP) with a cell density (e.g. McF). Thus, a species-specific LUT is created to allow for estimation of cell density from an ORP signal. (B) Five different strains of six organisms (E. coli, K. pneumoniae, S. marcescens, K. oxytoca, C. freundii, K. aerogenes) were cultured and bacterial suspensions of ~ 1 × 107 CFU/mL were prepared. These were grown at 37 °C and the ORP curves are shown here. (C) Species-specific algorithms were applied to the ORP signal curves to determine the predicted time at which an equivalent 0.5 McF was reached (time to eMcF). This was plotted as a function of the bacterial species and ranged from 64 min (one strain of K. pneumoniae) to 150 min (one strain of K. oxytoca). The variability in time to eMcF is due to different starting concentrations of bacteria as well as different growth rates for each strain. (D) The eMcF plated colony count data is plotted for all six bacterial species tested. These counts demonstrate a tight control of eMcF cell density as 28 of 30 eMcFs fell within ± 0.2log of our target concentration (1.5 × 108 CFU/ml).
