Development of an Intelligent System for Diagnosing Rumen Acidosis in Cows. Part 2: Computer Implementation

Vladimir V. Pobedinskiy; Andrey A. Pobedinskiy; Grigory A. Iovlev

doi:10.23947/2949-4826-2026-25-1-26-33

Development of an Intelligent System for Diagnosing Rumen Acidosis in Cows. Part 2: Computer Implementation

Vladimir V. Pobedinskiy, Andrey A. Pobedinskiy, Grigory A. Iovlev

https://doi.org/10.23947/2949-4826-2026-25-1-26-33

EDN: GHJLEA

Full Text:

PDF (Eng) PDF (Rus) HTML HTML (Rus) XML XML (Rus)

Generate QR code

Contents

Scroll to:

Abstract

Introduction. In the first part of the study, the relevance of improving the methods of rumen acidosis diagnostics in cows based on the intelligent systems was substantiated and the use of fuzzy set theory as a tool for such systems was proposed. The structure of the hierarchical-type intelligent system was developed, and formalization of the problem was completed. In the second part of the study, computer implementation of the fuzzy-logic-based intelligent system for diagnosing acidosis was completed using Xcos tool of Scilab software, and a computer application was developed in the Scilab+Scinotes environment as its practical implementation.

Materials and Methods. The study was conducted at Ural State Agrarian University and Northern Trans-Ural State Agricultural University from 2022 to 2025. The intelligent system techniques, fuzzy set theory and fuzzy logic methods, and Mamdani fuzzy inference system were used to conduct the study. A computer model for rumen pathology assessment was created based on the generalized smart system by means of Xcos tool of Scilab software. Computer implementation of the intelligent system was completed in the Scilab+Scinotes environment.

Results. For solving a problem of hierarchical-type intelligent system, the production rule bases, which included various combinations of diagnostic parameters and respective degrees of rumen acidosis pathology, were developed. Synthesis of fuzzy systems was performed using SciFLT tool of Scilab software. A generalized model of an intelligent system for diagnosing rumen acidosis in cows was developed using Xcos tool of Scilab software. Computer implementation of the intelligent diagnostic system was completed in Scilab software with embedded Scinotes text editor.

Discussion and Conclusion. The model of an intelligent system proposed by the authors is innovative and can be recommended for practical implementation into the expert advisory systems, for automation of veterinary workstations, and for using in modern veterinary telemedicine technologies.

Keywords

rumen acidosis in cows, rumen acidosis pathology, diagnostics, intelligent diagnostic system, fuzzy set theory, computer implementation

For citations:

Pobedinskiy V.V., Pobedinskiy A.A., Iovlev G.A. Development of an Intelligent System for Diagnosing Rumen Acidosis in Cows. Part 2: Computer Implementation. Russian Journal of Veterinary Pathology. 2026;25(1):26-33. https://doi.org/10.23947/2949-4826-2026-25-1-26-33. EDN: GHJLEA

Introduction. Ensuring sustainable development of animal husbandry is a key objective stipulated in the State Program for the Development of Agriculture in the Russian Federation [1]. Thus, the primary objective of the industry recognized at the government level for the nearest future is annual growth of cattle population and reduction of its mortality¹.

Rumen acidosis is one of the most widespread diseases in cattle. It is called “the prime cause underlying all health problems in cows” [2] and “the most acute problem of any herd, both the high-yielding or ordinary one” [3]. All this justifies the necessity to improve the diagnostics of acidosis at any stage of the disease. For this purpose, we have created an intelligent system for diagnosing rumen acidosis in cows. At the first stage of the study [4], formalization of the tasks was carried out based on the concept of fuzzy logic: a set of diagnostic parameters for assessing the degree of acidosis pathology was justified; the structure of the hierarchical-type intelligent system was developed; the problem was formalized using the fuzzy set theory methods. The objective of the second part of the study is to complete computer implementation of the earlier formalized fuzzy-logic-based intelligent system for assessing the rumen acidosis pathology in cows.

To achieve this objective, the following tasks were defined:

1) development of the rule bases to solve a problem of hierarchical-type intelligent system;

2) synthesis of fuzzy systems using SciFLT tool of Scilab software;

3) development of a generalized model of an intelligent system for diagnosing rumen acidosis using Xcos tool of Scilab software;

4) computer implementation of the intelligent diagnostic system in Scilab software with embedded Scinotes text editor.

Materials and Methods. The study was conducted at Ural State Agrarian University (Ekaterinburg) and Northern Trans-Ural State Agricultural University (Tyumen) from 2022 to 2025. Methods of veterinary medicine, the intelligent system techniques, fuzzy set theory and fuzzy logic methods, along with Mamdani fuzzy inference system were used to conduct the study. Synthesis of fuzzy systems was performed using SciFLT for Scilab software (Scilab Consortium, France). A computer model for rumen pathology assessment was created based on the generalized smart system by means of Xcos tool of Scilab software. Computer implementation of the intelligent system was completed in the Scilab+Scinotes environment.

Research Results

1. Development of the production rule bases. Various combinations of diagnostic parameters and corresponding to them degrees of acidosis pathology were taken to develop the rule bases. The degree of pathology will vary from “Permanent” to “Acute”, and the linguistic variable will take values from “Min” (the animal is healthy) to “Max” (acute phase).

Significant dependencies of the disease degree from the pH of rumen contents (pH ) and Fat content in milk (F) should be determined according to the rules of fuzzy set theory (“If A = B and C = D and ... then m_i = n_j and ...”) and be based on the known data [5–22].

The developed fuzzy production rule bases for inferencing the function Y₁₂ = f(pH, F), the function Y₃₄ = f(Pulse Rate, Respiratory Rate), and the resulting function for the Degree of Acidosis Pathology DAP = f(Y₁₂, Y₃₄) are presented in Tables 1–3.

Table 1

Rule base for inferencing the function Y₁₂ = f(pH, F)

Values of the linguistic variable “pH”	Values of the output fuzzy subsets “pH – Fat content, Y₁₂” upon changing the fuzzy function “Fat content, F”
Values of the linguistic variable “pH”	Minimum, Min	Low, L	Average, Av	High, H	Maximum, Мах
Highest Acidity, HighestAc	Max	Max	H	Av	L
Moderate Acidity, ModAc	Max	H	Av	L	L
Neutral	H	Av	L	Min	Min
Moderate Alkalinity, ModAlk	Av	L	L	Min	Min
Highest Alkalinity, HighestAlk	L	Min	Min	Min	Min

Table 2

Rule base for inferencing the function Y₃₄ = f(PR, RR)

Values of the linguistic variable “Pulse Rate, PR”	Values of the output fuzzy subsets “Respiration – Pulse, Y₃₄” upon changing the fuzzy function “Respiratory Rate, RR”
Values of the linguistic variable “Pulse Rate, PR”	Minimum, Min	Low, L	Average, Av	High, H	Maximum, Мах
Minimum, Min	Min	Min	Min	L	Av
Low, L	L	L	L	Av	H
Average, Av	L	Av	Av	H	Мах
High, H	Av	Av	H	Мах	Мах
Maximum, Мах	Av	H	Мах	Мах	Мах

Table 3

Rule base for inferencing the function DAP = f(Y₁₂, Y₃₄) = f(pH, F, PR, RR)

Values of the linguistic variable “pH – Fat content,Y₁₂”	Values of the output fuzzy subsets “DAP” upon changing the fuzzy function “Respiration–Pulse, Y₃₄”
Values of the linguistic variable “pH – Fat content,Y₁₂”	Minimum, Min	Low, L	Average, Av	High, H	Maximum, Мах
Minimum, Min	Min	Min	Min	L	L
Low, L	Min	Min	L	Av	Av
Average, Av	Min	L	Av	H	H
High, H	L	Av	H	H	Мах
Maximum, Мах	Av	H	Мах	Мах	Мах

2. Synthesis of fuzzy models of the problem. Fuzzy inference operations and synthesis of fuzzy models of the problem were performed using the rule bases. The SciFLT Editor toolbox for Scilab software was used to implement the developed formal formulation of the fuzzy inference problem, which is part of the intelligent system structure². Fuzzy inference for obtaining the resulting membership functions is shown in Fig. 1a–c. In accordance with the fuzzy inference methodology³, on the example of the function Y₁₂ = f(pH, F), the operations were performed in the following order:

1) fuzzification of problem variables (Fig. 1a–b);

2) development of a rule base (Fig. 1c).

Fig. 1. Fuzzy inference operations by means of SciFLT toolbox: a) pH and F variables; b) Y₁₂ and Y₃₄; c) fuzzy inference rule base for the function Y₁₂ = f(pH, F)

Note: Russian and Latin symbols in Fig. 1 с) correspond to the English equivalents as follows:

Кн – HighestAc

Кср – ModAc

Нейтраль – Neutral

Щср – ModAlk,

Щн – HighestAlk

SZ – F

Мин – Min

М – L

Ср – Av

В – H

After completing fuzzy inference and defuzzification, the resulting functions Y₁₂ = f(pH, F); Y₃₄ = f(PR, RR) and DAP = f(Y₁₂,Y₃₄) were obtained. They are shown graphically in Fig. 2.

Fig. 2. Resulting functions: a — Y₁₂ = f(pH, F); b — Y₃₄ = f(PR, RR); c — DAP = f(Y₁₂,Y₃₄)

3. Development of a generalized model of an intelligent diagnostic system. Figure 3 demonstrates a model of an intelligent system created by means of Xcos toolbox, which uses Constant visual blocks for inputting initial data and Mux multiplexers for generating a data sequence
vector and transmitting data to Controller blocks. These blocks address the corresponding fuzzy systems to obtain fuzzy inference results. The calculation results are displayed on a virtual electronic display.

Fig. 3. Model of an intelligent system created by means of Xcos toolbox

4. Computer implementation of the intelligent diagnostic system. To be used in practice, the model requires a user interface. Scilab+Scinotes were used for its development. The main visual interface forms are shown in Figure4. They provide the information about the program, documentation, input/output data, and recommendations for treating rumen acidosis depending on its pathology degree.

Fig. 4. The main user interface forms of the program: a — home form; b — data input/output and program operation form

Discussion and Conclusion. The Artificial Intelligence undoubtedly has great potential for development in the agro-industrial complex. The intelligent system for diagnosing rumen acidosis pathology in cows proposed by the authors is innovative and can be implemented into the expert advisory systems, be used for automation of workplaces of the veterinarians and in modern veterinary telemedicine technologies or for other purposes.

The adequate work of the system is an issue of particular importance, due to impossibility to apply the traditional, standard approaches to it. Unlike other models, intelligent rule-based systems, i.e., fuzzy systems and neural networks, get trained throughout their entire existence cycle [23]. This means that upon changing external conditions (e.g., methods of treatment, pharmacological agents, or the occurrence of side effects), new sets of rules will be incorporated into the fuzzy production rule bases, and the operation of intelligent system will be automatically adjusted. If this procedure is automated by means of software rather than performed manually, the system becomes self-learning.

The standard basic treatment protocols have been inserted as a template into the process of forming treatment recommendations. Possibility for a veterinarian to adjust and clarify the treatment is obligatory foreseen, as the software remains an advisory tool for a veterinarian to make a final decision.

Summing up the study results, it can be concluded that the proposed intelligent system is capable of quite accurate diagnostics of acidosis pathology and is versatile due to its ability to improve the diagnostic process (by adding new rules to the rule bases or adding other diagnostic parameters) without requiring any changes to the program code.

1. Strategy for the Development of Agro-Industrial and Fishery Complexes of the Russian Federation for the Period up to 2030. Approved by the Resolution of the Government of the RF No. 2567-r of September 8, 2022.

2. Scilab. URL: https://www.scilab.org (accessed: 19.06.2024).

3. Piegat A. Fuzzy Modeling and Control: with 96 Tables. Heidelber, New York: Physica Publ.; 2001. 760 p.

References

1. Amerkhanov HA. The Role and Place of the Livestock Industry in Ensuring Food Security in the Russian Federation. (In Russ.) URL: https://ran-szv.ru/index.php/doklads/amerkhanov-150524 (accessed: 29.01.2026).

2. Samolovov AA. Rumen Acidosis – The Cause of All Cow Health Problems. Production Disease. Novosi-birsk; 2016. 61 p. (In Russ.)

3. Tregubov VI. Methods for the Elimination and Prevention of Acidosis in Cows. (In Russ.) URL: https://agbz.ru/articles/metody-ustraneniya-i-preduprezhdeniya-atsidoza-u-korov/ (accessed: 11.02.2026).

4. Pobedinskiy VV, Pobedinskiy AA, Iovlev GA. Development of an Intelligent System for Diagnosing Ru-men Acidosis in Cows. Part 1: Formalization of Tasks. Russian Journal of Veterinary Pathology. 2025;24(4):17–26. (In Russ.) https://doi.org/10.23947/2949-4826-2025-24-4-17-26

5. Bakirov B, Ruzikulov NB, Khaitov B, Abdurasulov AKh. Group Prevention of Rumen Acidosis in Cows. Journal of Osh State University. Agriculture: Agronomy, Veterinary and Zootechnics. 2023;(4(5)):50–56. (In Russ.) https://doi.org/10.52754/16948696_2023_4_7

6. Vladimirov VE, Kirsanov VV, Pavkin DYu. Study of the Rumen's pH and Temperature for Milk Cows af-ter Calving Acidosis Diagnosis. Vestnik vserossiiskogo nauchno-issledovatel'skogo instituta mekhani-zatsii zhivotnovodstva (Bulletin of the All-Russian Research Institute of Animal Husbandry Mechaniza-tion). 2019;(4(36)):196–199. (In Russ.)

7. Voronov DV, Bober YuN. Indicators pH of the Content of Rumen in the Cows, Patients with Acidosis, with Different Methods of Obtaining the Sample. Scientific Notes of the Vitebsk State Awarded with the Badge of Honor Academy of Veterinary Medicine. 2017; 53(3):18–21. (In Russ.)

8. Lugovoi MM, Azarnova TO, Podolnikov VE, Lugovaya IS. Importance of Maintaining a Hydrogen Index in the Cows for Prevention of Metabolic Disorders and for an Increase in Milk Productivity. Agrarnaya Rossiya (Agrarian Russia). 2019;(12):3–7. (In Russ.) https://doi.org/10.30906/1999-5636-2019-12-3-7

9. Pavkin DYu, Vladimirov FE. Diagnosis of Subacute Rumen Acidosis in Cows in the Postpartum Period when Using Digital Technologies. Head of Animal Breeding. 2020;(12(209)):47–52. (In Russ.)

10. Pobedinsky AA. Pobedinsky VV. A Method for Monitoring Bee Hives and Bee Populations. Izvestia Orenburg State Agrarian University (Bulletin of the Orenburg State Agrarian University). 2023;(6(104)):198–204. (In Russ.)

11. Pobedinsky AA. Remote Monitoring Determining Productivity Laying Hens in a Private Household. Izvestia Orenburg State Agrarian University (Bulletin of the Orenburg State Agrarian University). 2022;(1(93)):127–130. (In Russ.) https://doi.org/10.37670/2073-0853-2022-93-1-127-130

12. Ryzhkova GF, Evglevsky AA, Evglevskaya EP, Minenkov NA. The Redistribution of Electrolytes be-tween the Erythrocytes and Plasma of Cows Blood at the Violation of Acid-Base Balance (Acidosis of the Rumen). Vestnik Kurskoi gosudarstvennoi sel'skokhozyaistvennoi akademii (Bulletin of Kursk State Agri-cultural Academy). 2018;(4):136–139. (In Russ.)

13. Elenshleger AA, Solovyeva VV. Clinical and Morphological Blood Parameters in Rumen Acidosis in Dairy Cows. Bulletin of Altai State Agricultural University. 2016;(6(140)):112–115. (In Russ.)

14. Abarghuei MJ, Rouzbehan Y, Salem AZМ, Zamiri МJ. Nutrient Digestion, Ruminal Fermentation and Performance of Dairy Cows Fed Pomegranate Peel Extract. Livestock Science. 2013;157(2–3):452–461.

15. AlZahal O, Kebreab E, France J, Froetschel M, McBrideet BW. Ruminal Temperature May Aid in the De-tection of Subacute Ruminal Acidosis. Journal of Dairy Science. 2008;91(1):202–207. https://doi.org/10.3168/jds.2007-0535

16. Antanaitis R, Juozaitiene V, Malasauskiene D, Televicius M., Urbutis M. Biomarkers from Automatic Milking System as an Indicator of Subclinical Acidosis and Subclinical Ketosis in Fresh Dairy Cows. Polish Journal of Veterinary Science. 2019;22(4):685–693. https://doi.org/10.24425/pjvs.2019.129981

17. Asqarov SS, Yunusov XB, Roziqulov NB. Qo‘zilar Dispepsiyasining Klinik Belgilari va Ularning Etiopa-togenetik Asoslari. Veterinariya meditsinasi. 2023;8:18-19.

18. Smith BP, Van Metre DC, Pusterla N (Eds). Large Animal Internal Medicine. 6th Edition. USA: Elsevier; 2020. 1874. https://doi.org/10.1016/C2016-0-01788-6

19. Chen L, Liu Sh, Wang H, Wang М, Lihuai Yu. Relative Significances of pH and Substrate Starch Level to Roles of Streptococcus Bovis S1 in Rumen Acidosis. AMB Express. 2016;6(1):80. https://doi.org/10.1186/s13568-016-0248-2

20. Chen L, Luo Y, Wang Н, Liu Sh, Shen Y, Wang M. Effects of Glucose and Starch on Lactate Production by Newly Isolated Streptococcus Bovis S1 from Saanen Goats. Applied and Environmental Microbiology. 2016;82(19):5982–5989. https://doi.org/10.1128/AEM.01994-16

21. Millen DD, De Arrigoni MB, Pacheco RDL (Eds). Rumenology. Cham: Springer; 2016. 314 p. https://doi.org/10.1007/978-3-319-30533-2

22. McSweeney CS, Mackie RI (Eds). Improving Rumen Function: Burleigh Dodds Series in Agricultural Science. Cambridge: Burleigh Dodds Science Publishing; 2020. 862 p.

23. The Era of Self-Learning: How Self-Learning AI is Transforming Industries. URL: https://rfidunion.com/ru/information/self-learning-ai-will-create-a-new-self-era.html (accessed: 11.02.2026)

About the Authors

Vladimir V. Pobedinskiy

Ural State Forest Engineering University; Ural State Agrarian University
Russian Federation

Vladimir V. Pobedinskiy, Professor, Dr.Sci. (Engineering), Head of the Management in Technical Systems and Innovative Technologies Department; Professor of the Service of Transport and Technological Machines and Equipment for the Agro-Industrial Complex Department,

Ekaterinburg.

Web of Science ResearcherID: G-3245-2018.

Scopus ID: 57210947239.

Andrey A. Pobedinskiy

Tyumen State University
Russian Federation

Andrey A. Pobedinskiy, Associate Professor of the Forestry, Woodworking and Applied Mechanics Department, Agrarian Institute,

18, Roshchinskoye Shosse, Tyumen, 625003.

Web of Science ResearcherID: G-3777-2018.

Grigory A. Iovlev

Ural State Agrarian University
Russian Federation

Grigory A. Iovlev, Cand.Sci. (Economics), Head of the Service of Transport and Technological Machines and Equipment for Agro-Industrial Complex Department,

Ekaterinburg.

Web of Science ResearcherID: MSW-7499-2025.

Scopus ID: 57203821332.

Review

For citations:

JATS XML

This work is licensed under a Creative Commons Attribution 4.0 License.

ISSN 2949-4826 (Online)

Username
Password
	Remember me
Not a user? Register with this site Forgot your password?

User

Russian Journal of Veterinary Pathology

Development of an Intelligent System for Diagnosing Rumen Acidosis in Cows. Part 2: Computer Implementation

Full Text:

Abstract

Keywords

For citations:

References

About the Authors

Review

For citations:

Cookies policy