RESEARCH/R&D
Rodent malaria models serve as essential preclinical antimalarial drugs and vaccine testing tools. Evaluating these models requires manually counting parasite-infected red blood cells (RBCs), a time-consuming, often unreliable, and repetitive process. Along with researchers at the Malaria Research Institute of Johns Hopkins University, we have developed machine learning (ML) software to expedite such studies by automating the counting of Plasmodium-infected RBCs in rodents. Previous ML methods created by our group, designed to count P. falciparum-infected RBCs in humans, accurately measure parasitemia but need to be optimized to measure parasitemia in rodent models. We fine-tuned our ML model to target P. yoelii and P. berghei in mouse RBCs. Our improved algorithm reliably detected P. yoelii and P. berghei-infected RBCs across a wide parasitemia range (0.13-74.12%). Automated parasitemia measurements strongly correlated with manual estimates by expert parasitologists.
Malaria Screener R was built on top of an existing software called LabelImg. LabelImg is a graphical image annotation tool often used by ML researchers in object detection projects for data labeling, a process where human agents prepare the raw data for the training process by marking images with ground truth information. Malaria Screener R can analyze blood smear images with an embedded ML module, present the analysis results with color-coded bounding boxes around infected and uninfected RBCs, and export the results to an Excel spreadsheet with graphical representations. It also allows manual corrections of potentially mislabeled RBCs. Malaria Screener R is a stand-alone desktop application that can be used on Windows and Mac OS machines. It can be run directly by double-clicking the executable without any installation process. The software is easy to use, and no computational knowledge is required.
Extramural researchers: Sean Yanikaa, Nattawat Chaiyawongaa, Opeoluwa Adewale-Fasoroaa, Luciana Dinisaa, Ariel Lubonjab, Bowen Libb and Prakash Srinivasanaa
a Department of Molecular Microbiology and Immunology and Malaria Research Institute, Johns Hopkins School of Public Health, Baltimore, MD 21205
b Department of Computer Science, Johns Hopkins Whiting School of Engineering, Baltimore, MD 21218
View the Malaria Screener R License Agreement
Mac OS
If the information above does not resolve the issue. Please do not hesitate to contact us.