Intelligent module of data analysis for information systems based on artificial neural networks

About the authors:

Alexander A. Zakharov, Dr. Sci (Tech.), Professor, Secure Smart City Information Technologies Department, University of Tyumen; a.a.zakharov@utmn.ru

Eugene A. Olennikov, Cand. Sci. (Tech.), Associate Professor, Head of the department of Information Security, University of Tyumen; e.a.olennikov@utmn.ru

Tatyana I. Payusova, Senior Lecturer, Department of Information Security, Institute of Mathematics and Computer Sciences, Tyumen State University

Abstract:

The aim of this study is to develop the algorithm for determining optimal architecture of artificial neural network for the analysis of medical data based on an evolutionary algorithm, implemented in the framework of a cloud service, access to which is organized on the basis of the protocols of the Unified State Information System in health care of the Russian Federation.Neural network methods and evolutionary algorithms are heuristic optimization techniques. Operating principles of neural network models and genetic algorithms are based on the processes occurring in nature. Models of artificial neural networks can solve the problem of classification, prediction, regression. Genetic algorithms are used to solve optimization problems and modeling. The algorithm of determining the optimum neural network architecture based on the evolutionary algorithm for the medical data analysis is considered in the paper.The genetic algorithm and artificial neural network are equal in the above algorithm, so both methods are used simultaneously. The operators of the genetic algorithm are used for recombination candidate solutions obtained using the neural network model. The cloud service that implements the presented algorithm is developed in the course of this work.

References:

1.             Gil’manov A. A., Sherputovskij V. G., Hisamutdinov A. N. Medicinskie informacionnye sistemy. Ispol’zovanie informacionnyh medicinskih sistem v upravlenii lechebno-profilakticheskim uchrezhdeniem [Medical Information Systems. The Use of Information Systems in Medical Preventive Institution Management]. Kazan’: KGMU, 2011. (In Russian)

2.             Edinaja gosudarstvennaja informacionnaja sistema zdravoohranenija (EGISZ). http://tadviser.ru/a/279359 (In Russian)

3.             Zaharov A. A., Olennikov E. A., Pajusova T. I., Petelina T. I., Musihina N. A., Gapon L. I., Osipova I. V., Takkand A. G., Belosludceva O. E. Nauchnyj analiz dannyh v medicinskoj informacionnoj sisteme, na primere opredelenija faktorov, vlijajushhih na uroven’ S-reaktivnogo belka, s pomoshh'ju nejronnyh setej [Scientific Data Analysis in Medical Information System (Case Study of Determining the Factors Affecting the Level of C-reactive Protein Using Neural Networks)] // Vestnik TjumGU [Tyumen State University Herald]. 2014. No 7. Pp. 251-258. (In Russian)

4.             Martynov A.V., Kurejchik V. M. Gibridnyj algoritm reshenija zadachi kommivojazhera [Hybrid Approach for Travelling Salesman Problem] // Izvestija JuFU. Tehnicheskie nauki [IZVESTIYA SFedU. Engineering Sciences]. 2015. No 4 (165). http://cyberleninka.ru/article/n/gibridnyy-algoritm-resheniya-zadachi-kommivoyazhera
(In  Russian)

5.             Pajusova T. I. Algoritm opredelenija faktorov, vlijajushhih na uroven’ C-reaktivnogo belka u pacientov s ishemicheskoj bolezn'ju serdca, s pomoshh’ju nejronnyh setej [The Algorithm for Determining the Factors Affecting the Level of C-reactive Protein in Patients with Coronary Heart Disease with the Help of Neural Networks] // Matematicheskoe i informacionnoe modelirovanie [Mathematical and Information Modeling]: Collection of Research Papers. Tyumen: Izdatel’stvo TjumGU [Tyumen State University Publishing House]. 2015. Issue 14. Pp. 193-205. (In Russian)

6.             Proekt dorozhnoj karty po razvitiju EGISZ [Project of Rroad Map for the Development of Integrated National Information System of Health Service] // Portal operativnogo vzaimodejstvija uchastnikov EGISZ [Web-portal for Operational Coordination of Members of Integrated National Information System of Health Service]. http://portal.egisz.rosminzdrav.ru/files (In Russian)

7.             Rutkovskaja D., Pilin’skij M., Rutkovskij L. Nejronnye seti, geneticheskie algoritmy i nechetkie sistemy [Neural Networks, Genetic Algorithms and Fuzzy Systems]. M.: Gorjachaja linija –Telekom [Hot Line — Telecom], 2006. 452 p. (In Russian)

8.             Trebovanija k MIS po obespecheniju vzaimodejstvija s FJeR. Opisanie integracionnyh profilej [Requirements for Medical Information System to Ensure Interaction with the Federal E-registration Office. Description of Integration Profiles] // Portal operativnogo vzaimodejstvija uchastnikov EGISZ [Web-portal for Operational Coordination of Members of Integrated National Information System of Health Service]. http://portal.egisz.rosminzdrav.ru/files (In Russian)

9.             Jel’janov M. M., Baranov A. P. Medicinskie informacionnye tehnologii [Medical-information technology]: Catalogue. M.: CapitalPress, 2012. Issue 12. 320 p. (In Russian)

10.         Jahjaeva G. Je. Osnovy iskusstvennyh nejronnyh setej. Material iz kursa: Osnovy teorii nejronnyh setej [Basics of Artificial Neural Networks. Material from the Course: Fundamentals of Theory of Neural Networks] http://cogsys.ru/sites/default/files/library/NeuroNets.pdf (In Russian)

11.         Baxt W. G. Use of an artificial neural network for the diagnosis of myocardial infarction // Ann Intern Med. 1991. No 115. Pp. 843-848. Doc: 10.7326/0003-4819-115-11-843.

12.         Chern H. Y., Miikkulainen R. Cooperative Coevolution of Multi-Agent Systems // Technical Report AI01-287. The University of Texas at Austin, 2001.

13.         Ding W., Zhou L., Bao Y., Yang Y., Lu B. Autonomic nervous function and baroreflex sensitivity in hypertensive diabetic patients // Acta Cardiol. 2011. No 66. Pp. 465-470.

14.         Gomez F. J., Miikkulainen R. Active Guidance for a Finless Rocket using Neuroevolution // Proceedings of the Genetic Evolutionary Computation Conference (GECCO). 2003.

15.         Gomez F. J., Miikkulainen R. Incremental evolution of complex general behavior // Adaptive Behavior. 1997. No 5. Рp. 317-342.

16.         Harrison R. F., Kennedy R. L. Artificial neural network models for prediction of acute coronary syndromes using clinical data from the time of presentation // Ann Emerg Med. 2005. No 46. Pp. 431-439. Doc. 10.1016/j.annemergmed.2004.09.012.

17.         Ottenbacher K. J., Linn R. T., Smith P. M., Illig S. B., Mancuso M. Comparison of logistic regression and neural network analysis applied to predicting living setting after hip fracture // Ann Epidemiol. 2004. No 14. Pp. 551-559.

18.         Whitley D. Genetic Algorithms and Neural Networks // Genetic Algorithms in Engineering and Computer Science. John Wiley, 1995. Рp. 203-216.