Key points are not available for this paper at this time.
Artificial intelligence (AI) enhances biosensor design by efficiently processing and modeling environmental data. This study employs machine learning algorithms to optimize biosensor parameters for the detection of trace-level heavy metals in aquatic environments, utilizing enzymes, DNAzymes, and aptamers as recognition elements. Machine learning models, including decision trees, random forests, gradient boosting, ensemble neural networks, and GLMM,were trained on extensive laboratory datasets. Among these, the random forest model exhibited the highest predictive accuracy, achieving 71 % for the limit of detection (LOD), 75 % for the minimum concentration of linearity, and 62 % for the maximum concentration of linearity. The AI-driven approach not only enhances biosensor sensitivity but also reduces experimental time and costs, enabling more efficient environmental monitoring. Moreover, this strategy is adaptable for detecting a wide range of pollutants, including chemical fertilizers, emerging contaminants, and micropollutants in aquatic and soil systems. These findings represent a significant step toward next-generation biosensing technologies with potential applications in sustainable environmental monitoring.
Noorisafa et al. (Thu,) studied this question.