Near‐sensor computing‐assisted simultaneous viral antigen and antibody detection via integrated label‐free biosensors with microfluidics

Abstract

Abstract Precise diagnosis and immunity to viruses, such as severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) and Middle East respiratory syndrome coronavirus (MERS‐CoV) is achieved by the detection of the viral antigens and/or corresponding antibodies, respectively. However, a widely used antigen detection methods, such as polymerase chain reaction (PCR), are complex, expensive, and time‐consuming Furthermore, the antibody test that detects an asymptomatic infection and immunity is usually performed separately and exhibits relatively low accuracy. To achieve a simplified, rapid, and accurate diagnosis, we have demonstrated an indium gallium zinc oxide (IGZO)‐based biosensor field‐effect transistor (bio‐FET) that can simultaneously detect spike proteins and antibodies with a limit of detection (LOD) of 1 pg mL–1 and 200 ng mL–1, respectively using a single assay in less than 20 min by integrating microfluidic channels and artificial neural networks (ANNs). The near‐sensor ANN‐aided classification provides high diagnosis accuracy (>93%) with significantly reduced processing time (0.62%) and energy consumption (5.64%) compared to the software‐based ANN. We believe that the development of rapid and accurate diagnosis system for the viral antigens and antibodies detection will play a crucial role in preventing global viral outbreaks.

Keywords

• artificial intelligence
• biosensor
• edge‐computing
• integration
• microfluidic