Bulletin "Veterinary biotechnology"

  • Print

Veterynarna biotehnologija – Veterinary biotechnology, 2016, 29, 43-54 [in Ukrainian].

BERHILEVYCH O.M, e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it., KASYANCHUK V.V.. e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it., CHERNECKII I., e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.

Sumy State University

TEREKHYNA O.V., e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.

Sumy National Agrarian University

BERHILEVYCH O.O., e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.

State Scientific and Research Institute of Laboratory Diagnostics and Veterinary and Sanitary Expertise

PREDICTION OF NUMBER OF PSYHROTROPHIC MICROORGANISMS IN REFRIGERATED COW’S MIKL WITH USING ARTIFICIAL NEURAL NETWORKS

Introduction. Considered that animal products have to be controlled on microbiological hazards, because these products are most often contaminated with dangerous microorganisms during their production. To control microbiological hazards during food processing the HACCP system is used. For the quick functioning of the HACCP system the companies should develop and implement rapid methods for the control of microbial contamination of food. These methods can be implemented using computer models and programs that reflect the probable relationship between certain parameters of the product, conditions of storage (processing) and the number of microorganisms.

The goal of the work was to develop a method of prediction of psyhrotrophic microorganisms number in cooled cow’s milk using artificial neural networks.

Materials and methods. 550 samples of raw cow’s milk were collected from dairy farms of Kyiv, Sumy and Zaporizhia oblasts (Ukraine). For research we used milk of extra and higher class. All samples were investigated for total number of microorganisms and number of psyhrotrophic microorganisms using standard methods. Predicting method was developed using experimental models and a special artificial neural network with computer programs NeuroPro.

Results of research and discussion. Development of a method carried out in several stages. In the beginning was molded base their experimental data with model experiments. In experiment models it was determined relationship between the total number of microorganisms and psyhrotrophic microorganisms in raw milk, temperature and terms of storage. The next stage involved the input of the data to the artificial neural network for predicting using NeuroPro program that contains three input parameters (total number of microorganisms, temperature and term of storage milk) and one output parameter (predictable count of psyhrotrophic microorganisms).

Conclusions and prospects for further research. The method of predict the number of psyhrotrophic microorganisms in raw cow’s during storage at low temperatures, was developed. This method has a high degree of reliability (from 98.0% to 99.8%) and its average error (standard deviation) of 0.5% to 2.0%. Advantages of this method included: its speed, informatively and significantly reduction of the number of tests for microorganisms number detection. This method will replace real experiments on mathematical models that adequately reflect the most important regularities of the objects. Application of the proposed method will facilitate effective control in milk production and timely use of corrective measures.

Keywords: prediction, artificial neural networks, psyhrotrophic microorganisms, milk.

REFERENCES

  1. Berhilevych, A.N., Marchenko, A.N., Berhilevych, O.A. (2014). Vliyanye sovremennih texnologyj doenyia i psyxrotrofnyh mikroorganizmov na soderzhanie svobodnih zhirnykh kislot v moloke [Influence of modern milking technology and psyhrotrophyc microorganisms on the content of free fatty acids in milk]. Nauchnij czurnal “ScienceRise” – Scientific Journal “ScienceRise”, 5/4(5), 61-68 [in Russian].
  2. Berhilevych, O.M. (2008). Prognozuyucha mikrobiologiya yak osnovoa dlia bezpeky produktiv kharchuvannia [Predictive microbiology is the basis for food safety]. Naukovyi visnik Lvivskogo natsionalnogo universytetu veterinarnoyi meditsiny ta biotexnologij im. S.Z. Gzhytskogo – Scientific Bulletin of Lviv National University of Veterinary Medicine and Biotechnology named S.Z. Hzhytskoho, Vol.10, № 2 (37), 3-7 [in Ukrainian].
  3. Berhilevych, O.M., Kasyanchuk, V.V., Zasyekin, D.A., Alekseyev, O.M., & Berhilevych, O.O. e.a. Sposib prognozuvannya kilkosti bakterij Enterobacter sakazakii v syromu oxolodzhenomu moloci protyagom zberigannya [The method of predict the number Enterobacter sakazakii bacteria in raw milk during refrigerated storage]. Patent Ukrayiny, no. 60298.
  4. Berhilevych, O.M., Kasyanchuk, V.V., & Berhilevych, O.O. e.a. Sposib prognozuvannya kilkosti pcihrotrofnix mikoorgnismiv v moloci koroviachomy zbirnomu oholodczenomy z vicoristnniam shtuchnih nejronnix merehz [The method of predict the number psyhrotrofnyh microorganisms in milk cow collectively cooled using artificial neural networks]. Patent Ukrayiny nо 108873.
  5. Kasyanchuk, V.V., Berhilevych, O.M., Marchenko, A.M., & Sklyar, O.I. (2013) Rozrobka metodu prognozuvannya zakhvoryuvanosti koriv na subklinichnyi mastit [Development of a method of predicting the incidence of subclinical mastitis in cows]. Visnyik Sumskogo NAU – Bulletin of the Sumy NAU, № 9 (33), 152-156 [in Ukrainian].
  6. Kasyanchuk, V.V., Berhilevych, O.M., Gryshyna, E. A. & Teroxina, O. (2016). Kontrol vidpovidnosti mikrobiologichnim kriteriyam u virobnitstvi molokoproduktiv [Monitoring compliance with microbiological criteria in the production of milk]. Zhurnal «Molochna industriya» – Journal “Dairy industry”, 2, 32-35 [in Ukrainian].
  7. Dvoreczkyj, D.S., Dvoreczkyi, S.Y., Muratova, E.Y., Ermakov, A.A., & Dvoreczkyi, D.S. (2005). Kompyuternoe modelyrovanye botexnologycheskyx processov y system: [Computer modeling of biotechnological processes and systems]. Textbook. Tambov: Yzd-vo TGTU [in Russian].
  8. Moloko i molochni produkty`. Viznachennia kilkosti mikroorganizmiv. Metod pidraxuvannya kolonij za temperatury 30°C [Milk and dairy products. Determination of the number of microorganisms. The method of calculating the colonies at a temperature of 30°C]. (2005). National Standard of Ukraine (IDF 100В:1991, IDT). DSTU IDF 100В-2003 from 1st January 2005. Kyiv: Derzhspozhyvstandart Ukrai’ny [in Ukrainian].
  9. Moloko. Metod pidraxovuvannia kolonij psikhrotrofnykh mikroorganizmiv, shcho formuyut koloniyi za temperatury 6,5°C [Milk. The method of calculating psyhrotrophs colonies of microorganisms that form colonies at a temperature of 6.5°C]. (2008). National Standard of Ukraine (IDF 101:2005, IDT). DSTU IDF 101:2006 from 1st January 2008. Kyiv: Derzhspozhyvstandart Ukrai’ny [in Ukrainian].
  10. Moloko koroviache nezbirane. Vimogi pri zakupivli [Cows milk. Requirements for purchase]. (1998). National Standard of Ukraine. DSTU 3662-97: 2007 from 1st January 1998. Kyiv: Derzhspozhyvstandart Ukrai’ny [in Ukrainian].
  11. Berhilevych, O.M., Kasyanchuk, V.V., Gryshyna, E.A., & Teroxina, O.V. (2014). Osnovni mikrobiologichni riziki pri otsintsi sanitarno-hihienichnogo virobnitstva moloka na fermakh [The main microbiological risks in assessing the hygienic status of milk production on farms]. Visnyik Sumskogo NAU – Bulletin of Sumy NAU, 6 (35), 94-98 [in Ukrainian].
  12. Hajmeer, M.N. Basheer I.A., & Marsden J.L. (2000) New approach for modeling generalized microbial growth curves using artificial neural networks. J. Rapid Methods and Automation in Microbiol., 8, No 4, 265-284.
  13. Jeyamkondana S., Jayasa D.S., & Holleyb R.A. (2001). Microbial growth modelling with artificial neural networks. Int. J. Food Microbiol., 64, 343-354.

Download full text in PDF