Deep Learning Algorithms Enabled Visual Detection of Anthrax Biomarkers by Mn₃O₄ Nanozyme-Based Colorimetric Sensor...

Area of Science:

• * Nanomaterials and Sensor Technology
• * Biomarker Detection and Chemical Analysis
• * Artificial Intelligence in Diagnostics

Background:

• * Accurate detection of the anthrax biomarker 2,6-pyridine dicarboxylic acid (2,6-PDA) is crucial for public health.
• * Existing detection methods can be time-consuming and labor-intensive.
• * Development of rapid, sensitive, and selective sensor systems is needed.

Purpose of the Study:

• * To develop an innovative colorimetric sensor array (CSA) for detecting 2,6-PDA and its structural analogs.
• * To integrate advanced algorithms, including deep learning, for automated analysis and classification.
• * To enhance the speed, accuracy, and cost-effectiveness of biomarker detection.

Main Methods:

• * Fabrication of a CSA using phenylalanine-modified Mn3O4 nanozymes with tunable oxidase-like activity.
• * Catalysis of 3,3',5,5'-tetramethylbenzidine (TMB) oxidation by nanozymes, producing quantifiable color changes.
• * Application of multivariate statistical analysis and the YOLOv8 deep learning algorithm for data interpretation and classification.

Main Results:

• * The CSA successfully discriminated among 2,6-PDA and six structural analogs.
• * Achieved a low detection limit (LOD) of 0.015 ± 0.002 μM for 2,6-PDA.
• * The YOLOv8 algorithm demonstrated high performance (mAP 0.98-0.99) for automated detection and classification.

Conclusions:

• * The developed CSA integrated with advanced algorithms offers a rapid and accurate method for 2,6-PDA detection.
• * Automated analysis using YOLOv8 significantly reduces detection time and labor costs.
• * This approach provides a robust platform for real-world applications in complex environments, such as biological fluid analysis.