V.A. Al-Khaidri - Post-graduate Student, Department Biomedical and Electronic systems and technology, Vladimir State University named after A.&N. Stoletovs. E-mail: fawaz_tariq@mail.ru R.V. Isakov - Ph. D. (Eng.), Associate Professor, Department Biomedical and Electronic systems and technology, Vladimir State University named after A.&N. Stoletovs. E-mail: Isakov-RV@mail.ru L.T. Sushkova - Dr. Sc. (Eng.), Professor, Head of Department Biomedical and Electronic systems and technology, Vladimir State University named after A.&N. Stoletovs. E-mail: ludm@vlsu.ru

The human body is a complex dynamic system, comprising a number of subsystems and processes. Several physiological processes can be active at the same. Each of them generates a plurality of signals of different types. The appearance of the signals from the processes and systems that are not currently objects of study can be regarded as a physiological disturbance. This raises the problem of improving the quality and reliability of the information content of functional diagnostics in medicine. As is known, the most dangerous and widespread diseases are the cardiovascular disease (CVD). To diagnose the functional state of the heart electrocardiography is used. It-s a recording of the electrical activity of the heart. The presence of noise and distortion in electrocardiographic signal leads to misdiagnosis, wrong conclusions, and as a result, the patient is assigned to the wrong treatment. Consequently, we need to develop methods and algorithms for the evaluation and improvement of quality ECG signal in order to avoid the above error. Analysis of the literature on methods for the detection of noise and distortion in bioelectric signals shows that the greatest application have such methods as Independent component analysis, wavelet transform and artificial neural network (ANN). In this paper the methods of complexing of discrete wavelet transform (DWT) and ANN. This approach is based on the wavelet trans-form for the disclosure of the frequency content of the signal with the localization of the time. ANN performs the role of ECG classifier in terms of the presence of distortions and their classification on the basis of «suitable» or «unsuitable» for the further interpretation and formation of medical conclusion. Detail coefficients obtained by the decomposition electrocardiosignal are used is ANN input. In this work a database of ECG images distorted without them have been created. ANN has been trained in the MatLab. The results showed the effectiveness of the proposed approach. The advantage of this approach is that it allows to reduce the time and computing costs by reducing the amount of input data from the decomposition of the electrocardiographic signal into the detail and the approximate coefficients.

 

1. Rangajjjan R.M. Analiz biomedicinskikh signalov. Prakticheskijj podkhod / Per. s angl. pod red. A.P. Nemirko. M.: FIZMATLIT. 2007. 440 s.
2. Vsemirnaja organizacija zdravookhranenija. http://www.who.int/mediacentre/factsheets/fs317/ru/.
3. Arie C Maan, Erik W van Zwet, Sumche Man, Suzanne MM Oliveira-Martens, Martin J Schalij, Cees A Swenne. Assessment of Signal Quality and Electrode Placement in ECGs using a Reconstruction Matrix // Computing in Cardiology. 2011. V. 38. P. 289−292.
4. Kozjura A.V. Ocenka diagnosticheskojj znachimosti ehlektrokardiograficheskogo signala // Trudy KH Mezhdunar. nauchnojj konf. «Fizika i radioehlektronika v medicine i ehkologii». Vladimir. 2012. T. 1. S. 152−155.
5. Akay M. Wavelet Applications in Medicine // IEEE Spectrum. 1997. V. 34. № 5. P. 50−56.
6. David R Achanccaray, Marco A Meggiolaro. Detection of artifacts from EEG data using wavelet transform, high-order statistics and neural networks. RiodeJaneiro. Brasil.2010.
7. Al-KHajjdri V.A., Isakov R.V., Sushkova L.T. Obzor osnovnykh metodov obnaruzhenija artefaktov v biomedicinskikh signalakh // FREHMEH-2014. S. 379−382.
8. Ivanko E.O., Ivanushkina N.G., Sinekop JU.S. Mnogourovnevyjj analiz ehlektrokardiogramm dlja vyjavlenija pozdnikh potencialov predserdijj // EHlektronika i svjaz. Tematicheskijj vypusk «EHlektronika i nanotekhnologii». CH. 2. 2009.
9. Geppener V.V., Lejjtane M.JA., CHernichenko D.A. Primenenie wavelet-preobrazovanija dlja klassifikacii artefaktov v ehlektroehncefalograficheskikh signalakh // Mezhdunar. konf. po mjagkim vychislenijam i izmerenijam (SCM-2000). Sankt-Peterburgskijj gosudarstvennyjj ehlektrotekhnicheskijj universitet «LEHTI» im. V.I. Uljanova (Lenina).
10. Josy Joy, P.Manimegalai. Wavelet Based EMG Artifact Removal From ECG Signal // Journal of Engineering, Computers & Applied Sciences (JEC&AS). August 2013. V. 2. № 8.
11. Deepak Vala, Tanmay Pawar, Thakar V.K. Identification and Analysis of Motion Artifact in Ambulatory ECG Signal in 1‑D and 2‑D. International // Journal of Emerging Technology and Advanced Engineering. September 2013. V. 3. № 9.
12. Grigorev D.S., Spicyn V.G. Primenenie nejjronnojj seti i diskretnogo vejjvlet-preobrazovanija dlja analiza i klassifikacii ehlektrokardiogramm. Tomskijj politekhnicheskijj universitet. 2012.
13. N.A. Al-KHulejjdi., Isakov R.V., Sushkova L.T. Issledovanie metodov kodirovanija vykhodov iskusstvennykh nejjronnykh setejj pri klassifikacii variabelnosti serdechnogo ritma // Nejjrokompjutery: razrabotka, primenenie. 2013. № 6. S. 48−54.
14. P. Senthil Kumar, Arumuganathan R., Sivakumar K., and Vimal C. Removal of Ocular Artifacts in the EEG through Wavelet Transform without using an EOG Reference Channel // Int. J. Open Problems Compt. Math. December 2008. V. 1. № 3. S. 188−200.
15. Slepnev E.S. Primenenie iskusstvennykh nejjronnykh setejj dlja analiza predavarijjnykh situacijj turboagregatov ehlektrostancijj // Internet-zhurnal «Naukovedenie». Majj-Ijun 2014. Vyp. 3.
16. Hassoun M.H. Fundamentals of Artificial Neural Networks. TheMITPress. 1995. 511 p.
17. Lkrichevskii M. Intellektualnye metody v menedzhmente. SPb: Piter. 2005. 304 s.
18. Kalackaja L.V., Novikov V.A., Sadov V.S. Organizacija i obuchenie iskusstvennykh nejjronnykh setejj: Ucheb. posobie dlja studentov. Minsk. 2002.
19. Zaencev I.V. Nejjronnye seti: osnovnye modeli: Ucheb. posobie k kursu «Nejjronnye seti». Voronezhskijj gosudarstvennyjj universitet. Voronezh. 1999.
20. Al-KHajjdri V.A., Isakov R.V., Sushkova L.T. Issledovanie vlijanija vybora funkcijj aktivacii na ehffektivnost raboty mnogoslojjnogo perseptrona // Nejjrokopmjutery: razrabotka i primenenie. 2015. № 7. S. 60−66.
21. JAsnickijj L.N. Iskusstvennyjj intellekt: populjarnoe vvedenie dlja uchitelejj i shkolnikov. http://inf.1september.ru/view_article.php-ID=200902304.