Distinguishing the Hook Effect in Lateral Flow Sandwich Immunoassays Using Deep-Learning Algorithm

Abstract

Hook effect is now widely present in sandwich lateral flow assay reactions, which refers to the phenomenon of false-negative results due to inappropriate antigen-antibody ratios, and can greatly limit the true color development of test strip strips under high sample concentration conditions, thus limiting quantitative detection. In this study, we developed a novel deep learning-based discrimination algorithm to accurately distinguish whether the strips are affected by the hook effect, which not only saves cost and manpower, but also clears the way for subsequent immunochromatographic quantification.