Machine Learning and Experimental Design-Driven Fluorescence Detection of Metronidazole in Food Samples Using Mold-Derived Carbon Quantum Dots

Carbon quantum dots (CQDs) were synthesized using mold as a green carbon source and characterized by a combination of techniques including fluorescence spectroscopy, Fourier-transform infrared spectroscopy (FTIR), Scanning Electron Microscopy (SEM), Energy-Dispersive X-ray Spectroscopy (EDX), elemental mapping, and line scan analysis. The CQDs showed strong fluorescence emission suitable for sensitive detection of the antibiotic metronidazole based on fluorescence quenching. After optimizing various parameters influencing sensor performance through experimental design, a linear detection ranges from 97.4 to 3215.3 µM with a detection limit of 50.5 µM was established. Machine learning algorithms were applied to enhance the accuracy and reliability of metronidazole quantification. This novel, eco-friendly sensor platform offers effective antibiotic monitoring for biomedical and environmental applications.