Oblique Back-Illumination Microscopy Enables Phase-Gradient Optical Sensing of Malaria-Induced Intracellular Heterogeneity

Abstract

Malaria remains a major global health burden, with diagnosis often relying on ex vivo analysis of infected red blood cells (RBCs), which incompletely samples key stages of parasite development. In particular, early ring-stage parasites exhibit subtle structural perturbations, while a significant fraction of parasite burden─including immature gametocytes─is sequestered within the microvasculature and is not readily captured in peripheral blood smears. Here, we present a label-free imaging and analysis framework that integrates oblique back-illumination microscopy (OBM) with data-driven classification to detect malaria infection by sensing parasite-induced intracellular heterogeneity. Unlike stained brightfield microscopy and quantitative phase imaging (QPI) methods that depend on interferometric acquisition and phase reconstruction, OBM directly converts intracellular refractive-index gradients into measurable intensity variations through noninterferometric phase-gradient contrast. By exploiting multiply scattered illumination and asymmetric detection, OBM enhances sensitivity to parasite-induced refractive-index gradients associated with Plasmodium falciparum development, capturing morphological complexity across the intraerythrocytic lifecycle. This enables the discrimination of uninfected cells, early ring-stage infections, and mature gametocytes. Fourier-domain analysis further reveals infection-stage-dependent redistribution of spatial-frequency content, providing a physically interpretable description of progressive intracellular remodeling during parasite development. Using synchronized in vitro cultures with brightfield validation, we demonstrate automated classification of infected RBCs with accuracies of 95% for binary classification and 82% for multiclass classification, with consistent precision, recall, and F1-scores across repeated randomized data splits. Together, these results establish OBM as a scalable, label-free sensing strategy that leverages intracellular heterogeneity as a diagnostic signal while providing access to stage-resolved optical phenotypes that remain difficult to interrogate with existing approaches. By building on prior studies of in vivo phase-gradient imaging, this framework provides a pathway toward integrating OBM with real-time analysis for noninvasive and field-deployable malaria diagnostics, particularly where conventional ex vivo measurements are limited by parasite sequestration.