Portable Electrochemiluminescence Microarray Sensor for Machine Learning-Assisted Quantitative Analysis of...

Area of Science:

• Analytical Chemistry
• Food Science
• Biotechnology

Background:

• Zearalenone (ZEN) is an estrogenic mycotoxin from Fusarium fungi, prevalent in cereals like wheat, rice, and maize.
• ZEN exposure poses risks, causing endocrine disruption and reproductive issues, necessitating reliable detection methods.

Purpose of the Study:

• To develop a portable, image-based electrochemiluminescence (ECL) platform for quantitative analysis of zearalenone (ZEN).
• To integrate a micropillar electrode design with machine learning for enhanced ZEN detection.

Main Methods:

• Fabrication of a portable ECL platform using luminol-functionalized Zr-based metal-organic frameworks (Zr@Lu-MOFs) on micropillar electrodes.
• Smartphone-based image acquisition of ECL signals, converted to RGB data for analysis.
• A two-step machine learning workflow involving K-nearest neighbor (KNN) classification and Gaussian process regression (GPR) for quantification.

Main Results:

• The platform achieved sensitive ZEN quantification across a wide linear range (0.0001–100 ng/mL) with a low limit of detection (LOD) of 0.23 pg/mL.
• Demonstrated high signal uniformity, reproducibility, and reliable performance in real food samples.
• The image-driven analytical strategy proved effective for scalable and portable analysis.

Conclusions:

• Integrating ECL microarrays with image-level data analysis offers an effective approach for developing scalable and portable analytical platforms.
• This method provides a promising tool for efficient and reliable food safety monitoring, particularly for mycotoxin detection.