Improving in Situ Electrode Calibration with Principal Component Regression for Fast-Scan Cyclic Voltammetry

Area of Science:

• Neuroscience
• Analytical Chemistry
• Biomedical Engineering

Background:

• Fast-scan cyclic voltammetry (FSCV) with carbon-fiber microelectrodes is vital for in vivo neurotransmitter detection, especially dopamine.
• Electrode calibration is crucial but challenging due to factors affecting sensitivity in neural tissue.
• Current in vitro calibration is standard, while in situ methods exist but can be improved.

Purpose of the Study:

• To enhance the accuracy and transferability of in situ electrode calibration for FSCV.
• To develop a novel calibration strategy using principal component regression (PCR) to analyze background waveform shape.
• To compare the performance of PCR-based calibration against the total-background-current strategy.

Main Methods:

• Utilized principal component regression (PCR) to model electrode sensitivity based on background waveform shape.
• Employed leave-one-out cross-validation for performance estimation and comparison.
• Applied multivariate quality-control statistics to validate the strategy with in vivo data.

Main Results:

• The PCR strategy demonstrated significantly superior predictive performance compared to the total-background-current method.
• Developed calibration models using PCR were successfully transferred across independent laboratories.
• Multivariate quality-control statistics confirmed the strategy's applicability to real-world in vivo FSCV data.

Conclusions:

• Principal component regression offers a more accurate and transferable approach for in situ FSCV electrode calibration.
• This advanced method improves the interpretation of in vivo electrochemical data, particularly for dopamine.
• Adoption of PCR-based calibration is recommended for advancing FSCV applications in neuroscience research.