Rational Design of High-Performance Nanozymes via Machine Learning for Smartphone-Based Quercetin Sensing

Quercetin (QR) has received widespread attention for its diverse biological activities. Therefore, establishing a rapid, accurate, and portable detection method is essential for safeguarding food safety and ensuring proper consumption. Herein, an efficient detection strategy was developed based on the peroxidase (POD)-like catalytic activity of the Fe 3 O 4 nanozymes. Machine learning (ML) algorithms were utilized for the rational design and optimization of Fe 3 O 4 nanozymes. Remarkably, Fe 3 O 4 nanozymes with an average size of 10 nm (E30) possessed superior catalytic activity. Moreover, a colorimetric sensor based on E30 was constructed to detect the QR. This assay presents a desirable linear relationship between the absorbance intensity and the concentrations of QR from 1 × 10 –8 to 5 × 10 –5 M with a detection limit (LOD) of 98 nM. In addition, a smartphone-assisted portable colorimetric platform was constructed to achieve a linear relationship ( R 2 = 0.9965) between the R–B signal and the concentration of QR. The method was also used to determine QR in chrysanthemum extract samples, and good recoveries and low relative standard deviations were obtained, showing great potential for the detection of active ingredients in Chinese herbal medicines.