The method of neuron weight vector initial values selection in Kohonen network

• Autorzy: Chandzlik S.
• Języki publikacji: EN

Abstrakty

EN

Diagnosing of morbid conditions by means of automatic tools supported by computers is a significant and often used element in modern medicine. Some examples of these tools are automatic conclusion-making units of Parotec System for Windows (PSW). In the initial period of PSW system implementation, the units were used for recognition of orthopaedic diseases on the basis of the patient's walk and posture [15,17]. Subsequently, many additional options have been implemented, which have been used for purposes of diagnosing neurological diseases [1,2,3,9,12]. During automatic classification of diseases the additional units use elements of neural networks. The vectors based on normalised diagnostic measures [3] are inputs of the units. The measurements describe a patient's posture condition, his walk and overloads occurring on his feet. The Counter-Propagation (CP), two-layer network has been used in one of the automatic conclusion-making units. During CP network activity, we can see not only supervised but unsupervised learning processes as well. This is a characteristic feature of the CP network. The initial steps of the CP network learning process are very important, because the success of the network training process depends on them to a great extent. Therefore, a new method of weight vector initial values selection was proposed. The efficiency of the method was compared with classical methods. The results were very satisfactory. Owing to the proposed method, the time of the network training process as well as the mean-square error and the classification error was reduced. The research has been carried out using clinical cases of some neurological diseases: Parkinson's Disease, left-lateral hemiparesis and right-lateral hemiparesis after ischemic stroke. The measurements, which were made on a control group of patients without any neurological diseases, were the reference for these diagnostic classes.

Słowa kluczowe

EN

neural networks; Kohonen network; neurological diseases diagnosis; Parkinson disease; hemiparesis after ischemic stroke

• Wydawca: University of Silesia, Institute of Informatics, Computer Systems Department
• Czasopismo: Journal of Medical Informatics & Technologies
• Rocznik: 2006
• Tom: Vol. 10
• Strony: 189--197
• Opis fizyczny: Bibliogr. 17 poz., fig., tab.

Twórcy

autor

Chandzlik S.

• University of Silesia, Institute of Informatics, Dept. of Computer Systems, Sosnowiec, Poland

Bibliografia

• [1] Chandzlik S., Kopicera K.: Experiments with neural network parameters – selection for foot abnormalities recognition, Journal of Medical Informatics & Technologies. Vol. 5, pp: CS-71 – CS-78. ISBN 83-909517-2-7, 2000.
• [2] Chandzlik S., Piecha J.: A patient walk-data-record modelling using a spline interpolation method. Journal of Medical Informatics & Technologies. Vol. 3, pp: MIT-153 – MIT-160. ISSN 1642-6037, 2002.
• [3] Chandzlik S., Piecha J.: The body balance measures for neurological disease estimation and classification. Journal of Medical Informatics & Technologies, Vol. 6, pp: IT-87 – IT-94, 2003.
• [4] Chandzlik S., Piecha J.: the gait characteristic data spectrum extraction. Proc. 4th Inter. Conf. on Computer Recognition System CORES’05, Vol. 18, pp. 493 – 501, 2005.
• [5] Floater M.S.: Parameterization and smooth approximation of surface triangulations. Comp. Aided Geom. Des., Vol. 14, pp: 231-250, 1997.
• [6] Hecht-Nielsen R.: Counterpropagation networks. Applied Optics, Vol. 26, pp: 4979-4984, 1987.
• [7] Hecht-Nielsen R.: Applications of counterpropagation networks. Neural Networks, Vol. 1, pp: 131-139, 1988.
• [8] Kopicera K., Piecha J.: The fault analysis made by PSW data recorder for neurological disease classification, Journal of Medical Informatics & Technologies, Vol. 4, pp: SN-10 – SN-13, 2002.
• [9] Kopicera K., Piecha J., Zyguła J.: The neural networks in diagnostics support for PSW system. Proc. of Int. Conference ASIS’99, pp. 113-118, Krnov, 1999.
• [10] Levin D., Nadler E.: Convexity preserving interpolation by algebraic curves and surfaces. Numerical Algorithms, Vol. 9, pp: 113-139, 1995.
• [11] Osowski S.: Sieci neuronowe w ujęciu algorytmicznym. WNT, Warszawa, 1996.
• [12] Piecha J.: The neural network conclusion-making system for foot abnormality recognition. Proceedings of IMACS World Congress, Lausanne, Switzerland, August 2000.
• [13] Piecha J., Kopicera K.: The conclusion making method using pathology classifiers, Proc. on KOSYR’01, pp: 29-33, 2001.
• [14] Piecha J., Zyguła J.: PC visual interface for orthopaedic expertise, Proc. of Int. Conference, pp. 162-167, 1995.
• [15] Piecha J., Zyguła J., Łyczak J., Gaździk T., Proksa J.: The advanced measuring system for orthopaedic pathologies diagnostics using a static and dynamic footprints, Chirurgia narządów ruchu i ortopedia polska vol. LXI 1996, suplement 3B, pp.119-124. (in Polish)
• [16] Tadeusiewicz R.: Sieci neuronowe. Akademicka Oficyna Wydawnicza RM, Warszawa, 1993.
• [17] Zyguła J.: Przetwarzanie danych pomiarowych dla systemu wnioskowania o patologiach w obszarze stopy. Rozprawa doktorska, Gliwice 1997. (in Polish)