Machine-learning assisted multiplex detection of catecholamine neurotransmitters with a colorimetric sensor array

Area of Science:

• * Analytical Chemistry
• * Nanotechnology
• * Biomarker Detection

Background:

• * Catecholamine neurotransmitters (CNs) are crucial biomarkers for neurological disorders.
• * Simultaneous monitoring of CNs is vital for accurate clinical diagnosis and treatment.
• * Existing methods for CNs detection face challenges in sensitivity and specificity.

Purpose of the Study:

• * To develop a high-performance colorimetric artificial tongue for multiplex detection of CNs.
• * To utilize gold nanoparticle aggregation and machine learning for sensitive and selective CN detection.
• * To validate the sensor array's performance in complex biological matrices like human urine.

Main Methods:

• * Fabrication of a colorimetric sensor array using gold nanoparticles (AuNPs).
• * Exploitation of pH-dependent AuNP aggregation triggered by varying acidity constants of CNs.
• * Application of machine learning algorithms (Linear Discriminant Analysis and Partial Least Squares Regression) for classification and quantification.

Main Results:

• * The sensor array generated unique fingerprint patterns for different CNs based on AuNP aggregation.
• * Accurate classification and quantification of dopamine (DA), epinephrine (EP), norepinephrine (NEP), and levodopa (LD) were achieved.
• * Excellent analytical figures of merit were obtained, with linear ranges from 0.1–70 μM and high R² values (≥0.99).
• * The sensor array demonstrated feasibility in human urine samples with low limits of detection (LODs).

Conclusions:

• * The developed colorimetric artificial tongue offers a promising platform for simultaneous, precise, and accurate determination of CNs.
• * This technology has significant potential for clinical diagnosis and monitoring of neurological illnesses.
• * The integration of nanomaterials and machine learning provides a powerful approach for complex biomarker analysis.