Machine Learning-Assisted Surface Ligand Engineering Strategy for Enhanced Sensitivity of Immunoassay Platform

Nanozymes hold great promise in point-of-care (POC) diagnostics. However, their lower catalytic activity has significantly limited their clinical applications. In this study, we report a surface engineering strategy based on charge-transfer ligand modulation to develop a trimetallic nanozyme (D-PtPdOs) with superior peroxidase-like activity, and integrate it with machine learning (ML) algorithms to enable ultrasensitive and intelligent detection of Pseudomonas aeruginosa (P. aeruginosa) in immunoassay platforms. By leveraging ligand-induced charge transfer, we precisely tuned the surface electron density of the nanozyme. The resulting D-PtPdOs nanozyme exhibits extraordinary catalytic activity, significantly outperforming natural horseradish peroxidase (HRP). Density functional theory (DFT) calculations reveal that d-histidine modification enhanced the adsorption capacity for hydrogen peroxide (H₂O₂) and lowered the activation energy barrier, thereby drastically increasing the maximum reaction rate (V_max). This research establishes a versatile surface ligand engineering paradigm, offering a novel design framework to overcome the catalytic bottlenecks inherent in nanozymes. Due to its superior catalytic activity, the D-PtPdOs was successfully integrated into enzyme-linked immunosorbent assay (ELISA) and lateral flow immunoassays (LFIA) platforms for P. aeruginosa detection, achieving sensitivity enhancements of 14.58-fold and 250-fold compared with conventional HRP-ELISA and AuNPs-LFIA, respectively. Furthermore, by incorporating ML algorithms, the platform enables high-precision classification and quantitative prediction of P. aeruginosa infection levels in complex human blood samples, effectively mitigating signal uncertainties caused by matrix interference. This work establishes a foundation for the deep integration of immunoassays with intelligent software and portable devices, significantly advancing the development of smart point-of-care diagnostics.