Sean Yanik, Hang Yu, Nattawat Chaiyawong, Opeoluwa Adewale-Fasoro, Luciana Ribeiro Dinis, Ravi Kumar Narayanasamy, Elizabeth C Lee, Ariel Lubonja, Bowen Li, Stefan Jaeger, Prakash Srinivasan
{"title":"Application of Machine Learning in a Rodent Malaria Model for Rapid, Accurate, and Consistent Parasite Counts.","authors":"Sean Yanik, Hang Yu, Nattawat Chaiyawong, Opeoluwa Adewale-Fasoro, Luciana Ribeiro Dinis, Ravi Kumar Narayanasamy, Elizabeth C Lee, Ariel Lubonja, Bowen Li, Stefan Jaeger, Prakash Srinivasan","doi":"10.4269/ajtmh.24-0135","DOIUrl":null,"url":null,"abstract":"<p><p>Rodent malaria models serve as important preclinical antimalarial and vaccine testing tools. Evaluating treatment outcomes in these models often requires manually counting parasite-infected red blood cells (iRBCs), a time-consuming process, which can be inconsistent between individuals and laboratories. We have developed an easy-to-use machine learning (ML)-based software, Malaria Screener R, to expedite and standardize such studies by automating the counting of Plasmodium iRBCs in rodents. This software can process Giemsa-stained blood smear images captured by any camera-equipped microscope. It features an intuitive graphical user interface that facilitates image processing and visualization of the results. The software has been developed as a desktop application that processes images on standard Windows and MacOS computers. A previous ML model created by the authors designed to count Plasmodium falciparum-infected human RBCs did not perform well counting Plasmodium-infected mouse RBCs. We leveraged that model by loading the pretrained weights and training the algorithm with newly collected data to target Plasmodium yoelii- and Plasmodium berghei-infected mouse RBCs. This new model reliably measured both P. yoelii and P. berghei parasitemia (R2 = 0.9916). Additional rounds of training data to incorporate variances due to length of Giemsa staining and type of microscopes, etc., have produced a generalizable model, meeting WHO competency level 1 for the subcategory of parasite counting using independent microscopes. Reliable, automated analyses of blood-stage parasitemia will facilitate rapid and consistent evaluation of novel vaccines and antimalarials across laboratories in an easily accessible in vivo malaria model.</p>","PeriodicalId":7752,"journal":{"name":"American Journal of Tropical Medicine and Hygiene","volume":null,"pages":null},"PeriodicalIF":1.9000,"publicationDate":"2024-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"American Journal of Tropical Medicine and Hygiene","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.4269/ajtmh.24-0135","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"PUBLIC, ENVIRONMENTAL & OCCUPATIONAL HEALTH","Score":null,"Total":0}
引用次数: 0
Abstract
Rodent malaria models serve as important preclinical antimalarial and vaccine testing tools. Evaluating treatment outcomes in these models often requires manually counting parasite-infected red blood cells (iRBCs), a time-consuming process, which can be inconsistent between individuals and laboratories. We have developed an easy-to-use machine learning (ML)-based software, Malaria Screener R, to expedite and standardize such studies by automating the counting of Plasmodium iRBCs in rodents. This software can process Giemsa-stained blood smear images captured by any camera-equipped microscope. It features an intuitive graphical user interface that facilitates image processing and visualization of the results. The software has been developed as a desktop application that processes images on standard Windows and MacOS computers. A previous ML model created by the authors designed to count Plasmodium falciparum-infected human RBCs did not perform well counting Plasmodium-infected mouse RBCs. We leveraged that model by loading the pretrained weights and training the algorithm with newly collected data to target Plasmodium yoelii- and Plasmodium berghei-infected mouse RBCs. This new model reliably measured both P. yoelii and P. berghei parasitemia (R2 = 0.9916). Additional rounds of training data to incorporate variances due to length of Giemsa staining and type of microscopes, etc., have produced a generalizable model, meeting WHO competency level 1 for the subcategory of parasite counting using independent microscopes. Reliable, automated analyses of blood-stage parasitemia will facilitate rapid and consistent evaluation of novel vaccines and antimalarials across laboratories in an easily accessible in vivo malaria model.
期刊介绍:
The American Journal of Tropical Medicine and Hygiene, established in 1921, is published monthly by the American Society of Tropical Medicine and Hygiene. It is among the top-ranked tropical medicine journals in the world publishing original scientific articles and the latest science covering new research with an emphasis on population, clinical and laboratory science and the application of technology in the fields of tropical medicine, parasitology, immunology, infectious diseases, epidemiology, basic and molecular biology, virology and international medicine.
The Journal publishes unsolicited peer-reviewed manuscripts, review articles, short reports, images in Clinical Tropical Medicine, case studies, reports on the efficacy of new drugs and methods of treatment, prevention and control methodologies,new testing methods and equipment, book reports and Letters to the Editor. Topics range from applied epidemiology in such relevant areas as AIDS to the molecular biology of vaccine development.
The Journal is of interest to epidemiologists, parasitologists, virologists, clinicians, entomologists and public health officials who are concerned with health issues of the tropics, developing nations and emerging infectious diseases. Major granting institutions including philanthropic and governmental institutions active in the public health field, and medical and scientific libraries throughout the world purchase the Journal.
Two or more supplements to the Journal on topics of special interest are published annually. These supplements represent comprehensive and multidisciplinary discussions of issues of concern to tropical disease specialists and health issues of developing countries