Fig. 3: Correlation of MO signals with parasite density as determined by expert LM.

a MO signals as a function of parasite density for different species, as determined by expert LM. P. falciparum mono-infections without gametocytes are represented by red circles, P. vivax mono-infections by blue diamonds and P.f. + P.v. mixed infections by green triangles. Samples with gametocytes in P. falciparum mono-infections (P.f. + gam.) are represented by orange squares. The larger, solid-colored symbols show median MO signals and median parasite densities for representative parasite density ranges, with 95% CIs as error bars. The smaller, transparent symbols show the individual test results. The number of biologically independent samples examined over three independent MO experiments is indicated in parentheses. The typical error of the MO signal for a given sample (measured in triplicates) is 11% (individual error bars are not indicated). The diagonal dashed line with a slope equal to 1 represents linear proportionality of the MO signal with parasite density. The dashed horizontal black line represents the cut-off value (MO = 4.19 mV/V) determined by the ROC analysis for all LM positive samples (see Table 2). The dashed horizontal gray line is the median MO background signal (0.47 mV/V) determined in malaria naïve volunteers, which provides a significantly lower cut-off value. b MO signals of patient samples with undetectable parasites by expert LM (n = 627). Also indicated are the median, 1.67 (95% CI: 1.57–1.78), and the IQR (1.06–2.89) of the MO signals.