Mn,Ce-CDs-Loaded MIL-53(Fe) Nanozymes Multisubstrate Sensor Array Integrated with Multiple Machine Learning...

Area of Science:

• Nanomaterials and Sensor Technology
• Biomedical Diagnostics
• Analytical Chemistry

Background:

• Conventional antioxidant detection methods suffer from interference, instrument dependency, and unreliable results.
• Accurate antioxidant detection is critical for biomedical and food safety applications.

Purpose of the Study:

• To develop an innovative nanozyme-based multisubstrate sensor array integrated with machine learning for enhanced antioxidant detection.
• To overcome the limitations of existing methods, including structural similarity interference and unreliable outcomes.

Main Methods:

• A novel nanocomposite of manganese- and cerium-codoped carbon dots (Mn,Ce-CDs) immobilized on MIL-53(Fe) was synthesized.
• The nanozyme catalyzed four chromogenic substrates, generating distinct multicolor fingerprint patterns for multidimensional signal output.
• Seven machine learning algorithms, including LDA, HCA, ANN, KNN, SVM, DT, and NB, were employed, with novel LDA-DT and LDA-NB tandem algorithms developed for improved accuracy.

Main Results:

• The nanozyme exhibited enhanced superoxide anion-mediated oxidase-like activity, enabling sensitive catalysis of chromogenic substrates.
• The sensor array successfully differentiated eight structurally analogous antioxidants with a minimum identification concentration of 10 nM.
• Tandem machine learning algorithms (LDA-DT and LDA-NB) achieved 100% classification accuracy after dimensionality reduction.
• The system accurately quantified and identified multiple antioxidants in complex matrices such as serum, urine, cell lysates, bacterial cultures, and food samples.

Conclusions:

• The developed nanozyme-based sensor array integrated with machine learning offers a robust and highly accurate platform for antioxidant detection.
• This innovative approach overcomes key challenges in conventional methods, providing reliable detection in diverse and complex biological and food matrices.
• The study demonstrates the potential of nanozyme technology and machine learning for advancing analytical capabilities in biomedical and food science.