Wyniki wyszukiwania - BazTech

1

Application of SOM in classification of EGG signals

Świerczyński Z., Guszkowski T.

Metrology and Measurement Systems

|

2006

|

Vol. 13, nr 3

279-287

EN

The report presents problems associated with computer aided gastric diagnosis. The subject of the study are electrogastrographic (EGG) signals (non-invasively measured electrical signals generated by the human stomach). The signals were digitally recorded and then parametrized, with linear autoregressive models (AR). The data and parametrization method used in the study were the same as used by the authors in the previous study; therefore here they are only shortly described. The sets of numbers, obtained by these means, were treated as information vectors, and classified with the Self Organizing Map (SOM) classifier. The structure and parameters of the algorithm used for classification of the parametrized EGG data are described. The final efficiency of the whole system (SOM classifier with the parametrization method applied), reaching 80%, is promising. It is similar to the results of other classifiers. The ways to improve the effectiveness are also outlined.

PL

Praca przedstawia problemy związane z komputerowo wspomaganym diagnozowaniem układu pokarmowego. Obiektem badań są tutaj sygnały elektrogastrograficzne - EGG (nieinwazyjnie mierzone sygnały elektryczne generowane przez żołądek człowieka). Sygnały te zostały zarejestrowane cyfrowo a następnie poddane parametryzacji przy pomocy liniowego modelu autoregresyjnego AR. Dane oraz metoda parametryzacji użyta w przedstawionych badaniach zostały opisane w poprzednich pracach autorów, więc tutaj ujęte są jedynie w zarysie. Zestawy liczb otrzymane w wyniku parametryzacji potraktowane zostały jako wektor parametrów i sklasyfikowane przy pomocy klasyfikatora opartego na samoorganizujących się mapach (SOM). W pracy przedstawiono strukturę i parametry użytego algorytmu. Ostateczna skuteczność całego systemu (tj. klasyfikatora SOM oraz zastosowanej metody parametryzacji) wyniosła 80%, co jest wynikiem obiecującym i bardzo podobnym do tych jakie osiągnięto przy zastosowaniu innych metod klasyfikacji. Praca przedstawia również zarys metod poprawy efektywności opisanego systemu.

2

Application of neural network and genetic algorithms in computer aided gastric diagnostic system

Świerczyński Z., Zagańczyk A.

Biocybernetics and Biomedical Engineering

|

2005

|

Vol. 25, no. 1

49-58

EN

In the paper, computer aided stomach diagnosis problems are presented. The subject of the study is electrical signal generated by human stomach, called the electrogastrographic (EGG) signal. The non-invasively measured signals were subjected to parametrization and then neural network based classification. The parametrization was performed with one of the time series modelling methods, with linear autoregressive models (AR). A special feature of the presented methodology of classification is its hybrid approach. The idea of this specific combination is that a genetic algorithm is used as the evolutionary method of training of the neural network. The structure and parameters of the system (NEUROGEN v.02), used for classification of the parameterized EGG data, are described. The finally obtained effectiveness of the whole system (NEUROGEN v.02 with the parametrization method applied), amounting to 74%, is quite high and according to the authors' analysis, can be improved. A way of improvement of the effectiveness are also outlined in the summary.

3

Application of SVM in computer aided gastric diagnostic system

Świerczyński Z., Mazur J.

Biocybernetics and Biomedical Engineering

|

2004

|

Vol. 24, no. 4

19-30

EN

In the paper computer aided stomach diagnosis problems are presented. The subject of the study is electrical signal generated by human stomach and called electrogastrographic (EGG) signal. The non-invasively measured signals were subjected to parametrization, which was performed with one of the time series modelling methods, with linear autoregressive models (AR). Then the obtained sets of numbers were classified with the Support Vector Machine (SVM), which is a relatively new pattern recognition technique and is based on the idea of structural risk minimization, The structure and parameters of algorithm used for classification of the parameterized EGG data are described. The finally obtained effectiveness of the whole system (SVM with the parametrization method applied), amounting to 81%, is promising and, according to the authors' analysis can be improved. The ways of improving of the effectiveness are also outlined in the conclusions.