Details
| Original language | English |
|---|---|
| Pages (from-to) | 135-141 |
| Number of pages | 7 |
| Journal | Analytica chimica acta |
| Volume | 348 |
| Issue number | 1-3 |
| Publication status | Published - 20 Aug 1997 |
Abstract
A computational neural network based evaluation method is presented, which enables a reliable quantification of enzyme field effect transistor (EnFET) flow injection analysis (FIA) signals from samples with changing pH values. Two FIA systems, one for glucose and the other for urea determination, are employed to test the evaluation method. Measurement signals were obtained from samples with different glucose concentrations (3, 4, 5, 6 and 7 g/l) and urea concentrations (1, 1.25, 1.5, 1.75 and 2.0 g/l at various pH values (5.5, 5.75, 6.0, 6.25 and 6.5). These signals cannot be evaluated based on the peak height, width or integral. Using a large set of measuring signals for training the artificial neural network (12 samples, each measured fivefold (=60) signals) the error of analyte prediction from test signals are 3.2% and 2.5% for glucose and urea respectively. With a reduced training set of five measurement signals the error of prediction of the test set increases to 4.5% and 5.5% for glucose and urea respectively. In this investigation it will be demonstrated that computational neural networks are able to evaluate FIA signals, which cannot be evaluated reliably by FIA standard methods.
Keywords
- Biosensors, EnFET, Evaluation method, Flow injection analysis, Neural networks, pH independent, Sample matrix
ASJC Scopus subject areas
- Chemistry(all)
- Analytical Chemistry
- Biochemistry, Genetics and Molecular Biology(all)
- Biochemistry
- Environmental Science(all)
- Environmental Chemistry
- Chemistry(all)
- Spectroscopy
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
In: Analytica chimica acta, Vol. 348, No. 1-3, 20.08.1997, p. 135-141.
Research output: Contribution to journal › Article › Research › peer review
}
TY - JOUR
T1 - Computational neural networks for the evaluation of biosensor FIA measurements
AU - Hitzmann, B.
AU - Ritzka, A.
AU - Ulber, R.
AU - Scheper, T.
AU - Schügerl, K.
PY - 1997/8/20
Y1 - 1997/8/20
N2 - A computational neural network based evaluation method is presented, which enables a reliable quantification of enzyme field effect transistor (EnFET) flow injection analysis (FIA) signals from samples with changing pH values. Two FIA systems, one for glucose and the other for urea determination, are employed to test the evaluation method. Measurement signals were obtained from samples with different glucose concentrations (3, 4, 5, 6 and 7 g/l) and urea concentrations (1, 1.25, 1.5, 1.75 and 2.0 g/l at various pH values (5.5, 5.75, 6.0, 6.25 and 6.5). These signals cannot be evaluated based on the peak height, width or integral. Using a large set of measuring signals for training the artificial neural network (12 samples, each measured fivefold (=60) signals) the error of analyte prediction from test signals are 3.2% and 2.5% for glucose and urea respectively. With a reduced training set of five measurement signals the error of prediction of the test set increases to 4.5% and 5.5% for glucose and urea respectively. In this investigation it will be demonstrated that computational neural networks are able to evaluate FIA signals, which cannot be evaluated reliably by FIA standard methods.
AB - A computational neural network based evaluation method is presented, which enables a reliable quantification of enzyme field effect transistor (EnFET) flow injection analysis (FIA) signals from samples with changing pH values. Two FIA systems, one for glucose and the other for urea determination, are employed to test the evaluation method. Measurement signals were obtained from samples with different glucose concentrations (3, 4, 5, 6 and 7 g/l) and urea concentrations (1, 1.25, 1.5, 1.75 and 2.0 g/l at various pH values (5.5, 5.75, 6.0, 6.25 and 6.5). These signals cannot be evaluated based on the peak height, width or integral. Using a large set of measuring signals for training the artificial neural network (12 samples, each measured fivefold (=60) signals) the error of analyte prediction from test signals are 3.2% and 2.5% for glucose and urea respectively. With a reduced training set of five measurement signals the error of prediction of the test set increases to 4.5% and 5.5% for glucose and urea respectively. In this investigation it will be demonstrated that computational neural networks are able to evaluate FIA signals, which cannot be evaluated reliably by FIA standard methods.
KW - Biosensors
KW - EnFET
KW - Evaluation method
KW - Flow injection analysis
KW - Neural networks
KW - pH independent
KW - Sample matrix
UR - http://www.scopus.com/inward/record.url?scp=0030870951&partnerID=8YFLogxK
U2 - 10.1016/S0003-2670(97)00153-0
DO - 10.1016/S0003-2670(97)00153-0
M3 - Article
AN - SCOPUS:0030870951
VL - 348
SP - 135
EP - 141
JO - Analytica chimica acta
JF - Analytica chimica acta
SN - 0003-2670
IS - 1-3
ER -