Article or book title, author's surname, DOI

The synthesis technique of a neural network subsystem of cardiorhythmogram signal processing

О.В. Мельник1, С.В. Челебаев2, Ю.А. Челебаева3

research article UDC 615.47.03:616.12-073.96 Pages: 45-52 doi.org/10.18127/j15604136-201904-07

  1. 1 д.т.н., профессор, кафедра информационно-измерительной и биомедицинской техники, Рязанский государственный радиотехнический университет им. В.Ф. Уткина
  2. 2 к.т.н., доцент, кафедра автоматизированных систем управления, Рязанский государственный радиотехнический университет им. В.Ф. Уткина
  3. 3 аспирант, кафедра информационно-измерительной и биомедицинской техники, Рязанский государственный радиотехнический университет им. В.Ф. Уткина

Abstract

Task of the analysis of a cardio rhythm in real time is detection of early arrhythmias for the purpose of their treatment and prevention of life-endangering arrhythmias. Implementation of the neural network organization of systems in a basis of devices creation of cardiorhythmogram analysis allows to expand also significantly their functionality, including learnability of the neural network device on selection of required informative sign. It is necessary to develop the synthesis technique regulating the main development stages of a subsystem for the purpose of ordering of design process of a neural network subsystem of cardiorhythmogram signal processing.

Purpose − development of synthesis technique of a neural network subsystem of cardiorhythmogram signal processing.

The following problem definition of subsystem creation of cardiorhythmogram signal processing is accepted: it is necessary to create set of neural network structures of cardiorhythmogram signal processing at execution of restrictions for tolerated error and preset time of processing, having provided minimization of an indicator of costs of its subsequent hardware implementation on FPGAs.

The enlarged order of subsystem design of cardiorhythmogram signal processing on the basis of neural network technologies is offered.

Representation of the main types of neurons in logical basis is carried out, their classification is executed.

The synthesis technique is applied to network design of informative sign determination of constant bigeminal pulse.

Different arrhythmias can exist together and the developed technique allows to find this combination in parallel. Further complex interpretation of signs is planned. This approach to creation of control systems of a cardio rhythm can be in real time used as for monitoring of already diagnosed cardiovascular diseases, especially in chambers of an intensive care, and for prevention and early diagnostics of symptoms of arrhythmia at persons with high myocardial risk.

How to cite

Мельник О.В., Челебаев С.В., Челебаева Ю.А. Методика синтеза нейросетевой подсистемы обработки сигналов кардиоритмограммы // Биомедицинская радиоэлектроника. 2019. № 4. С. 45–52. DOI: https://doi.org/10.18127/j15604136-201904-07

References
  1. Bereznyj E.A., Rubin A.M., Utekhina G.A. Prakticheskaya kardioritmografiya M.: Neo. 2005. 140 c.
  2. Mel'nik O.V., CHelebaev S.V., CHelebaeva Yu.A. Analiz serdechnogo ritma v rezhime real'nogo vremeni na osnove iskusstvennyh nejronnyh setej // Biotekhnosfera. 2016. № 6. S. 33–39.
  3. Mel'nik O.V., CHelebaev S.V., CHelebaeva Yu.A. Avtomatizirovannyj analiz serdechnogo ritma v rezhime real'nogo vremeni // Biomedicinskaya radioelektronika. 2017. № 7. S. 45–52.
  4. Golovko V.A. Nejronnye seti: obuchenie, organizaciya i primenenie. M.: IPRZHR. 2001. 256 s.