METHOD AND SYSTEM FOR LABEL-FREE IMAGING AND CLASSIFICATION OF MALARIA PARASITES

METHOD AND SYSTEM FOR LABEL-FREE IMAGING AND CLASSIFICATION OF MALARIA PARASITES

Researchers at the CZ Biohub SF have developed an improved method of detecting and classifying malarial parasitemia in blood samples that is cost-efficient and readily able to be implemented in low-resource settings.

Accurate, timely, and cost-effective diagnostics are a crucial component of the clinical treatment of infectious diseases. Diagnosis of infections in low-resource settings remains a barrier to proper and expeditious treatment and contributes to preventable mortality. Malaria, caused by infection of red blood cells (RBCs) with the parasite P. Falciparum, is among the largest contributors to infectious-disease related mortality in these regions. Despite a prevalence of 228 million cases annually, little has changed in the malaria diagnostic process in over a century. The current gold standard of malarial parasite visualization is microscopic inspection of fixed and stained human blood smears. This process is labor intensive, skill-dependent, and has been shown to be particularly vulnerable to human error. Recent advances in deep learning models has made it possible to automate the analysis and quantification of images, including those generated from microscopes. In parallel, visible light microscopy remains a cost-effective, readily available tool that produces high-contrast, high-resolution images.

Stage of Research

The inventors establish a bespoke method for the quantification and high throughput characterization of P. falciparum parasitemia in human blood smears. Using a commercially available visible light microscope and the latest advances in deep learning, they demonstrate a highly sensitive and specific way to achieve automated and label-free classification of live, parasitized RBCs. Additionally, they show that they are able to accurately classify P. falciparum parasites in RBCs into their distinct life cycle stages including the ring stage, which is the dominant circulating form of the parasites in human RBCs and often the most challenging to detect using traditional blood smear methods. This method was validated using primary human blood samples as well as commercially available visible light microscopes, demonstrating its translational potential as a diagnostic in low resource settings.

Applications

Highly sensitive and specific analysis that combines visible light microscopy and deep learning to allow for automated, label free classification of healthy and P. falciparum infected RBCs.
Accurate identification life stages of the parasite P. falciparum in human blood samples.

Advantages

Label free analysis of clinical samples, reducing man hours necessary to process samples while improving accuracy
Cost effective solution that is able to be implemented in areas with the most need for this technology, namely low-resource settings.
Highly sensitive and specific identification of healthy vs infected RBCs as well as accurate classification of parasite stage (ring, trophozoite, schizont) in infected RBCs

Stage of Development

Research- in vitro

Publications

Lebel P, Dial R, Vemuri VNP, Garcia V, DeRisi J, Gómez-Sjöberg R. Label-free imaging and classification of live P. falciparum enables high performance parasitemia quantification without fixation or staining. PLoS Comput Biol. 2021 Aug 9;17(8):e1009257. doi: 10.1371/journal.pcbi.1009257. PMID: 34370724; PMCID: PMC8376094.

WO2022/047171