2026-10-12T00:06:03Z https://api.figshare.com/v2/oai

oai:figshare.com:article/33868796 2026-09-17T00:08:46Z category_146 category_915 category_7 category_12 category_21 category_46 category_931 portal_63 item_type_2 month_year_09_2026

Optical Diffraction Tomography and Interpretable Machine Learning Reveal Biophysical Signatures of Gametocyte-Stage Malaria in Red Blood Cells Lintong Wu (8563140) Anoushka Gupta (22336344) Piyush Raj (12295676) Abhai K. Tripathi (322101) Ishan Barman (330976) Space Science Physical Sciences not elsewhere classified Medicine Cell Biology Biotechnology Immunology Information Systems not elsewhere classified supervised vision transformer supervised models trained rapid diagnostic tests optical diffraction tomography future diagnostic workflows current diagnostic tools significantly reduced sphericity plasmodium falciparum infection differentiate infection stages results establish odt increased eccentricity compared free malaria staging including sphericity falciparum cultures using three transmission control throughput analysis synchronized p strongest predictors stained microscopy specific diagnosis resolved phenotyping refractive index real time rdts ), powerful modality potential applications multiclass classification limited resolution interpretable machine interpretable framework including giemsa extract data dry mass downstream analysis derived features culture monitoring complementary features clinical management classify malaria biophysical descriptors binary classification Accurate, stage-specific diagnosis of Plasmodium falciparum infection is critical for clinical management and transmission control of malaria. Current diagnostic tools, including Giemsa-stained microscopy and rapid diagnostic tests (RDTs), are constrained by their reliance on labeling, limited resolution, and an inability to quantify or differentiate infection stages in real time. Here, we present a label-free approach using optical diffraction tomography (ODT) combined with an interpretable machine-learning framework to characterize and classify malaria-infected red blood cells (RBCs) across ring and gametocyte stages. Using three-dimensional (3D) refractive index (RI) tomograms of RBCs from synchronized P. falciparum cultures, we extract physically interpretable morphological and biophysical descriptors, including sphericity, solidity, eccentricity, dry mass, and maximum RI. In addition, we extract data-driven image representations using a self-supervised vision transformer (ViT) and integrate these complementary features for downstream analysis. Gametocyte stage-infected RBCs exhibit distinct morphology, with significantly reduced sphericity and increased eccentricity compared to those of ring stage-infected or normal RBCs. Supervised models trained on the combined feature space achieve 88.3% accuracy in multiclass classification (normal, ring, gametocyte) and 98.1% accuracy in binary classification (normal vs infected). SHAP analysis reveals deep-learning-derived features as the strongest predictors for normal and ring-stage RBCs, with handcrafted features critically influencing gametocyte-stage classification, reflecting pronounced morphological changes in gametocyte-stage RBCs. Our results establish ODT as a powerful modality for quantitative, label-free malaria staging and demonstrate a generalizable and interpretable framework for stage-resolved phenotyping with potential applications in high-throughput analysis, culture monitoring, and future diagnostic workflows. 2026-09-16T00:00:00Z Dataset Media 10.1021/cbmi.6c00102.s003 https://figshare.com/articles/media/Optical_Diffraction_Tomography_and_Interpretable_Machine_Learning_Reveal_Biophysical_Signatures_of_Gametocyte-Stage_Malaria_in_Red_Blood_Cells/33868796 CC BY-NC 4.0