Deep learning of HIV field-based rapid tests

• Published in: Nature medicine Vol. 27; no. 7; pp. 1165 - 1170
• Main Authors: Turbé, Valérian, Herbst, Carina, Mngomezulu, Thobeka, Meshkinfamfard, Sepehr, Dlamini, Nondumiso, Mhlongo, Thembani, Smit, Theresa, Cherepanova, Valeriia, Shimada, Koki, Budd, Jobie, Arsenov, Nestor, Gray, Steven, Pillay, Deenan, Herbst, Kobus, Shahmanesh, Maryam, McKendry, Rachel A.
• Format: Journal Article
• Language: English
• Published: New York Nature Publishing Group US 01.07.2021; Nature Publishing Group
• Subjects: 692/699/255/1901; 706/134; AIDS Serodiagnosis - methods; Algorithms; Applications programs; Biomedical and Life Sciences; Biomedicine; Cancer Research; Deep Learning; Diagnosis; Disease control; Field study; HIV; HIV infection; HIV Infections - diagnosis; Human immunodeficiency virus; Humans; Image acquisition; Image classification; Income; Infectious Diseases; Learning algorithms; Letter; Low income groups; Machine learning; Medical imaging; Medical personnel; Medical tests; Metabolic Diseases; Mobile computing; Molecular Medicine; Neurosciences; Quality assurance; Rural Health Services - organization & administration; Sensitivity and Specificity; South Africa; Technology application; Time and Motion Studies; South Africa; United Kingdom

Although deep learning algorithms show increasing promise for disease diagnosis, their use with rapid diagnostic tests performed in the field has not been extensively tested. Here we use deep learning to classify images of rapid human immunodeficiency virus (HIV) tests acquired in rural South Africa...