Skip to main content
Fig. 2 | Malaria Journal

Fig. 2

From: Infrared spectroscopy coupled to cloud-based data management as a tool to diagnose malaria: a pilot study in a malaria-endemic country

Fig. 2

a Partial least squares discriminant analysis (PLS-DA) prediction plot showing the classification either malaria positive (< 0.5) or negative (> 0.5); spectra colour-coded malaria positive (red) or negative (green) by PCR. b Same as in a) except support vector machine (SVM) learning is used for the classification. c Receiver operating characteristic (ROC) curves showing the diagnostic of the PLS-DA and SVM classification. d ROC curve for data where samples were assigned positive- and negative, based on PCR versus randomized models. e Average spectra over the three spectral ranges used for PLS-DA classification. Superimposed is a colour code showing the regression loadings for malaria positive (“warm colours”) or negative (“cool colours”) classification for each absorbance value

Back to article page