AI-assisted smartphone-based colorimetric biosensor for visualized, rapid and sensitive detection of pathogenic...

Area of Science:

• Biomedical Engineering
• Biosensor Technology
• Artificial Intelligence in Diagnostics

Background:

• Accurate bacterial detection is crucial for combating infections and contamination.
• Novel biosensors offer promising solutions for rapid and sensitive pathogen identification.
• Existing methods often lack speed, sensitivity, or ease of use for point-of-care applications.

Purpose of the Study:

• To develop an artificial intelligence (AI)-assisted smartphone-based colorimetric biosensor for detecting pathogenic bacteria.
• To utilize hyaluronidase (HAase) as a target biomarker for bacterial identification.
• To achieve visualized, rapid, and sensitive bacterial detection with a low limit of detection.

Main Methods:

• Development of a dual-hydrogel system: hyaluronic acid (HA) hydrogel loaded with chlorophenol red-β-D-galactopyranoside (CPRG) and agar hydrogel loaded with β-galactosidase (β-gal).
• HAase secreted by bacteria degrades the HA hydrogel, releasing CPRG, which then reacts with β-gal to produce a colorimetric signal.
• Utilized a self-developed YOLOv5 algorithm for AI-powered analysis of colorimetric signals captured by a smartphone.

Main Results:

• The biosensor provides results within 60 minutes.
• Achieved an ultra-low limit of detection (LoD) of 10 CFU/mL.
• Demonstrated 100% sensitivity in evaluating clinical samples and successfully differentiated between Gram-positive and Gram-negative bacteria.

Conclusions:

• The developed AI-assisted smartphone biosensor enables rapid, sensitive, and visualized detection of pathogenic bacteria.
• The biosensor shows significant potential for broad biomedical applications, aligning with the