Rapid, cost-effective, sensitive, and point-of-care disease sensors have become highly sought since the COVID-19 pandemic. We present an approach that achieves the sensitivity of nucleic acid amplification tests in less time. It combines an agglutination assay, lensfree microscopy, and machine learning. Antibody-coated beads sandwich virus particles and agglutinate depending on the viral concentration. By identifying particularly low levels of agglutination, we achieve a 1270 copies/mL limit of detection, comparable to polymerase chain reaction tests. Readout is performed in a cost-effective and compact device. This approach can be sped up further and used to identify other diseases in the future.
|