A Neural Network Model in the Calibration of Glucose Sensor Based on the Immobilization of Glucose Oxidase into Polypyrrole Matrix

• Published in: Electroanalysis (New York, N.Y.) Vol. 16; no. 18; pp. 1542 - 1549
• Main Authors: Seker, S., Becerik, I.
• Format: Journal Article
• Language: English
• Published: Weinheim WILEY-VCH Verlag 01-09-2004; WILEY‐VCH Verlag
• Subjects: Biosensors; Calibration; D-Glucose; Direct and inverse modelling; Neural networks
• ISSN: 1040-0397; 1521-4109
• DOI: 10.1002/elan.200302974

The immobilization of enzymes on an electrode surface is of great importance in bioelectrochemistry. The entrapment of enzymes into a polymer matrix is simple and a speedy technique for the production of biosensors. This procedure of enzyme immobilization by electropolymerization has a great significance in fabrication of micro sensors in the preparation of multiplayer devices. In current study, glucose oxidase enzyme that is specific for the glucose determination was entrapped into polypyrrole matrix containing p‐benzoquinone in PIPES buffer and glucose sensitivity of the biosensor was investigated. Then, artificial neural network analysis was done for the nonlinear calibration plot. This implementation can be used for the sensor failure detection, as well. The estimation power of the neural network used in the direct and inverse calibration modelling was examined by statistical methods. It presented the good performance for the estimation power.