|Article title||CLASSIFICATION OF REALIZATIONS OF MEDICAL AND BIOLOGIC SIGNALS WITH USE OF THE METHOD OF NONLINEAR TRANSFORMATIONS|
|Authors||G.G. Galustov, E.E. Zavtur|
|Section||SECTION I. METHODS AND ALGORITHMS FOR SIGNAL PROCESSING|
|Month, Year||11, 2014 @en|
|Abstract||The quasi-heuristic approach giving solving rules quite comprehensible from the engineering point of view and realised without special technical efforts is considered. A problem formalisation of recognition is based on objectively existing generality objects properties of the same class, often defined by concept of "similarity". The hypothesis about compactness using in the work assumes adequacy «similarity" concepts of one class signals and their geometrical "affinity" shown in their association in one coupled subset in space of signs. Questions of the integrated description of medical and biologic signals on the basis of nonlinear transformation of initial sets for the purpose of shaping optimum from the put forward criterion positions of a solving rule are considered. The procedure of the operator determination of the nonlinear transformation is resulted. It allows to construct an optimum dividing surface in space of signs for what it is not required an aprioristic knowledge of density functions of signs classified signals. It can seem that using only the first and the second moments essentially reduces value of the considered solving rule compared, for example, Bayesian rule as the fullest information about signals is concluded in their multidimensional distributions. However in the majority of practical problems of recognition these distributions are a priori unknown and cannot be received with satisfactory accuracy because of limitation of training sample. Therefore, for allocation of the useful information from the point of view of recognition it is possible to use only such statistical characteristics which possess sufficient stability at their estimation on real training sample.|
|Keywords||Training sample; final statistics; signals; casual processes; the operator of transformation; probability density; criterion of the optimality; solving rule; probability of an error; reliability of an estimation; nonlinear transformation.|
|References||1. Galustov G.G. Otsenka pogreshnostey vychisleniya statisticheskikh kharakteristik, pri ispol'zovanii metoda stokhasticheskogo kodirovaniya sluchaynykh protsessov [Evaluation of
errors of calculation of statistical characteristics, using the method of stochastic coding of stochastic processes], Materialy X Mezhdunar. nauchno-prakt. konf. «Nauchnyy progress na rubezhe tysyacheletiy» [Materials of the X Intern. scientific-pract. conf. "Scientific progress on
the Millennium" (Prague, 22-30 May 2014)]. Prague, 2014, pp. 80-86.
2. Galustov G.G., Brovchenko S.P., Krasnobaev D.A., Potsykaylo A.A. Otsenka tochnosti preobrazovaniya kod-veroyatnost' pri modelirovanii artefaktnykh shumov [Evaluation of the accuracy of the conversion code in the simulation, the probability of artifact noise],
Telekommunikatsii [Telecommunications], 2013, No. 3, pp. 5-12.
3. Galustov G.G., Brovchenko S.P., Tarasenko A.V. Analiz vliyaniya izmeneniya dinamicheskogo diapazona signala na realizatsiyu reshayushchego pravila pri reshenii zadachi klassifikatsii [Analysis of the impact of changes in the dynamic range of the signal for the implementation of a decision rule for solving the problem of classification], Izvestiya YuFU. Tekhnicheskie nauki [Izvestiya SFedU. Engineering Sciences], 2013, No. 11 (148), pp. 32-37.
4. Galustov G.G., Krasnobaev D.A., Potsykaylo A.A. O postroenii statisticheskikh sistem raspoznavaniya po klasterizovannym vyborkam [On the construction of statistical pattern recognition systems in clustered samples], Materialy Vseros. nauch. konf. «Sovremennye issledovatel'skie i obrazovatel'nye tekhnologii» (Taganrog, 30 oktyabrya 2010 g.) [Materials All-Russia scientific conf. "Modern research and educational technology" (Taganrog, October
30, 2010)]. Taganrog, 2010, pp. 45-50.
5. Galustov G.G., Potsykaylo A.A., Krasnobaev D.A. Sintez reshayushchego pravila klassifikatora signalov pri neparametricheskoy apriornoy neopredelennosti [Synthesis of a solving rule classifier signals in nonparametric a aprioristic indeterminacy], Izvestiya YuFU. Tekhnicheskie nauki [Izvestiya SFedU. Engineering Sciences], 2011, No. 1 (114), pp. 78-84.
6. Zakharov S.M., Galustov G.G. Mikroprotsessornyy modul' predvaritel'noy obrabotki mediko-biologicheskikh signalov [Microprocessor module pretreatment biomedical signals], Voprosy
radioelektroniki [Questions electronics], 1988, Issue 2, pp. 187-197.
7. Galustov G.G. Klassifikator sluchaynykh signalov [Qualifier casual signals], Izvestiya SKNTs VSh. Tekhnicheskie nauki [News SKNTS VS. Engineering], 1984, No. 3, pp. 54-57.
8. Gladkiy V.S. Veroyatnostnye vychislitel'nye modeli [Probabilistic computational models]. Moscow: Nauka, 1973, 298 p.
9. Kiselev N.V., Sechkin V.A., Stavitskiy A.I. Postroenie optimal'noy razdelyayushchey ploskosti v zadache raspoznavaniya obrazov [Construction of an optimal dividing plane in the problem of pattern recognition], Avtomatizatsiya proizvodstva [Manufacture automation], 1974, Issue 1, pp. 72-77.
10. Gorelik A.L., Skripkin V.A. Metody raspoznavaniya: ucheb. posobie dlya vuzov [Recognition methods: the manual for high schools], 4th ed. Moscow: Vysshaya shkola, 2004, 261 p.
11. Tu Dzh., Gonsales R. Printsipy raspoznavaniya obrazov [Principles of pattern recognition]. Moscow: Mir, 1978, 411 p.
12. Fukunaga K. Vvedenie v statisticheskuyu teoriyu raspoznavaniya obrazov [Introduction to statistical pattern recognition theory]. Moscow: Nauka, 1979, 367 p.
13. Galustov G.G. Modelirovanie sluchaynykh protsessov i otsenivanie ikh statisticheskikh kharakteristik [Modelling of casual processes and estimation of their statistical characteristics].
Moscow: Radio i svyaz', 1999, 120 p.
14. Omel'chenko V.A. Raspoznavanie signalov po spektru v usloviyakh apriornoy neopredelennosti [Recognition of signals on a spectrum in the conditions of aprioristic indeterminacy]. Khar'kov: KhPI, 1979, 100 p.
15. Galustov G.G., Tsymbal V.G., Mikhalev M.V. Prinyatie resheniy v usloviyakh neopredelen-nosti [Decision-making in the conditions indeterminacy]. Moscow: Radio i svyaz', 2001, 196 p.