Search PubMed⌕ Search

Biomedical subjects

Frank Dieterle

Publications and source records attributed to Frank Dieterle.

5 recordsLinked to original sources
PubMed2004

Different approaches to multivariate calibration of nonlinear sensor data.

In this study, different approaches to the multivariate calibration of the vapors of two refrigerants are reported. As the relationships between the time-resolved sensor signals and the concentrations of the analytes are nonlinear, the widely used partial least-squares regression (PLS) fails. Therefore, different methods are used, which are known to be able to deal with nonlinearities present in data. First, the Box-Cox transformation, which transforms the dependent variables nonlinearly, was applied. The second approach, the implicit nonlinear PLS regression, tries to account for nonlinearities by introducing squared terms of the independent variables to the original independent variables. The third approach, quadratic PLS (QPLS), uses a nonlinear quadratic inner relationship for the model instead of a linear relationship such as PLS. Tree algorithms are also used, which split a nonlinear problem into smaller subproblems, which are modeled using linear methods or discrete values. Finally, neural networks are applied, which are able to model any relationship. Different special implementations, like genetic algorithms with neural networks and growing neural networks, are also used to prevent an overfitting. Among the fast and simpler algorithms, QPLS shows good results. Different implementations of neural networks show excellent results. Among the different implementations, the most sophisticated and computing-intensive algorithms (growing neural networks) show the best results. Thus, the optimal method for the data set presented is a compromise between quality of calibration and complexity of the algorithm.

Journal Article↗
PubMed2003

Multi-analyte assay for triazines using cross-reactive antibodies and neural networks.

A biosensor system based on total internal reflectance fluorescence (TIRF) was used to discriminate a mixture of the triazines atrazine and simazine. Only cross-reactive antibodies were available for these two analytes. The biosensor is fully automated and can be regenerated allowing several hundreds of measurements without any user input. Even a remote control for online monitoring in the field is possible. The multivariate calibration of the sensor signal was performed using artificial neural networks, as the relationship between the sensor signals and the concentration of the analytes is highly non-linear. For the development of a multi-analyte immunoassay consisting of two polyclonal antibodies with cross-reactivity to atrazine and simazine and different derivatives immobilised on the transducer surface, the binding characteristics between these substances like binding capacity and cross-reactivity were characterised. The examination of three different measurement procedures showed that a two-step measurement using only one antibody per step allows a quantification of both analytes in a mixture with limits of detection of 0.2 microg/l for atrazine and 0.3 microg/l for simazine. The biosensor is suitable for online monitoring in the field and remote control is possible.

Algorithms↗
PubMed2003

Urinary nucleosides as potential tumor markers evaluated by learning vector quantization.

Modified nucleosides were recently presented as potential tumor markers for breast cancer. The patterns of the levels of urinary nucleosides are different for tumor bearing individuals and for healthy individuals. Thus, a powerful pattern recognition method is needed. Although backpropagation (BP) neural networks are becoming increasingly common in medical literature for pattern recognition, it has been shown that often-superior methods exist like learning vector quantization (LVQ) and support vector machines (SVM). The aim of this feasibility study is to get an indication of the performance of urinary nucleoside levels evaluated by LVQ in contrast to the evaluation the popular BP and SVM networks. Urine samples were collected from female breast cancer patients and from healthy females. Twelve different ribonucleosides were isolated and quantified by a high performance liquid chromatography (HPLC) procedure. LVQ, SVM and BP networks were trained and the performance was evaluated by the classification of the test sets into the categories "cancer" and "healthy". All methods showed a good classification with a sensitivity ranging from 58.8 to 70.6% at a specificity of 88.4-94.2% for the test patterns. Although the classification performance of all methods is comparable, the LVQ implementations are superior in terms of more qualitative features: the results of LVQ networks are more reproducible, as the initialization is deterministic. The LVQ networks can be trained by unbalanced sizes of the different classes. LVQ networks are fast during training, need only few parameters adjusted for training and can be retrained by patterns of "local individuals". As at least some of these features play an important role in an implementation into a medical decision support system, it is recommended to use LVQ for an extended study.

Adult↗
PubMed2002

Discrimination of methanol and ethanol vapors by the use of a single optical sensor with a microporous sensitive layer.

The sorption of methanol and ethanol vapors by a microporous glassy polycarbonate is studied. The increase of the refractive index of the polymer during analyte sorption is measured by surface plasmon resonance. Both analytes are sorbed into the micropores of the polymer showing different diffusion kinetics. The sensor response during analyte exposure is subdivided into different time channels. By evaluating this additional data dimension by neural networks, a simultaneous multicomponent analysis of binary mixtures of ethanol and methanol vapors is possible using the sensor response of only one single sensor. A feature extraction results in an interpretable model and an improved prediction with errors of 2.0% for methanol and 2.4% for ethanol.

Journal Article↗