Minimum Minutes Machine-Learning Microfluidic Microbe Monitoring Method (M7).

Area of Science:

• Environmental Science
• Analytical Chemistry
• Biotechnology

Background:

• Viral disease outbreaks pose significant societal and economic risks.
• Detecting low-concentration, complex viral aerosols in air is challenging with traditional methods.
• Manual collection and re-detection are time-consuming and cumbersome.

Purpose of the Study:

• To develop a rapid and accurate method for detecting virus aerosols in the air.
• To overcome the limitations of traditional virus detection techniques.

Main Methods:

• Utilizing microfluidic inertial separation and spectroscopic analysis for aerosol detection.
• Designing a microfluidic chip based on inertial separation and laminar flow principles.
• Integrating a microfluidic chip with a composite spectrometer for dynamic spectral information capture.
• Employing machine learning for accurate classification of aerosol particles.

Main Results:

• Achieved an average separation efficiency of 95.99% for 2 μm particles using the microfluidic chip.
• The entire detection process took less than 30 minutes, significantly faster than PCR.
• The developed model demonstrated a high accuracy of 97.87% in identifying virus aerosols.
• Results are comparable to those obtained through traditional PCR detection methods.

Conclusions:

• The proposed method offers a rapid, accurate, and efficient approach for virus aerosol detection.
• Microfluidic separation combined with spectroscopic analysis and machine learning is a promising strategy for airborne pathogen surveillance.
• This technology has the potential to significantly improve early detection and response to viral outbreaks.