Automated, Deep Learning-Enabled Immunoassay Microfluidic Cartridge for Viral Pathogen Detection

Area of Science:

• Biotechnology
• Medical Diagnostics
• Microfluidics

Background:

• Conventional Enzyme-linked immunosorbent assays (ELISA) require specialized equipment and trained personnel, limiting their use in resource-limited or at-home settings.
• Point-of-care diagnostic technologies often lack the sensitivity and accuracy of laboratory-based methods.

Purpose of the Study:

• To develop a disposable, electricity-free microfluidic chip (VISTA) for rapid, automated immunoassay-based diagnostics at the point-of-care.
• To integrate VISTA with an AI-enabled smartphone application for simplified result interpretation.

Main Methods:

• VISTA utilizes pressure-driven microfluidics, magnetic beads, and platinum nanoparticles for a semiautomated immunoassay process.
• The device consolidates sample processing and reagent addition into a single cartridge.
• An AI-powered smartphone application with an adversarial neural network analyzes assay results from images.

Main Results:

• VISTA successfully detected SARS-CoV-2 N antigen and HCV core antigen in 52 patient samples within 45 minutes.
• The diagnostic sensitivity of VISTA is comparable to laboratory-based ELISAs.
• VISTA demonstrated superior sensitivity compared to existing point-of-care technologies, detecting viral loads below 10^4 copies/mL.

Conclusions:

• VISTA offers a sensitive, cost-effective, and user-friendly solution for point-of-care diagnostics, particularly in low-resource settings.
• The integration of microfluidics and AI enables accessible disease detection without specialized expertise or equipment.
• This technology has the potential to significantly improve diagnostic capabilities in diverse healthcare environments.