METHOD OF NEURONET EVALUATION OF INCIDENCE RATE FOR NECROBACILLOSIS IN CATTLE
Abstract
Based on retrospective epizootological data were created DBF4 databases on comprehensive epizootological data, feeds, nutritional values of diets and biochemical blood serum examinations in cattle. The master database is of integrated type; it contains indicators on 158 animal buildings of farms having a necrobacillosis problem. This database uses the following input parameters: data on sanitary state, length of stall, presence or absence of grids, physical exercise, total nutritional value of diet, diet components. The database on feeds was created as well. A particular database on nutritive values of diets according to input parameters was prepared. A neural network was formed with the help of database on biochemical blood serum values of cattle from various farms. Based on epizootological databases across farms were developed artificial neural networks using NeuroPro 0.25 to forecast morbidity of cattle with necrobacillosis. Results of forecasting incidence rate for necrobacillosis in cattle at 18 farms with known epizootological data are given obtained with the help of the neuroproject developed.
Keywords
информационная модель,
нейронная сеть,
некробактериоз,
прогнозирование заболеваемости,
факторы риска,
санитарное состояние,
эпизоотологические данные,
база данных,
нейропроект,
information model,
neural network,
necrobacillosis,
forecasting incidence rate,
risk factors,
sanitary state,
epizootological data,
database,
neuroproject
About the Authors
S. V. Lopatin
Institute of Experimental Veterinary Science of Siberia and the Far East, Russian Academy of Agricultural Sciences
Russian Federation
A. A. Samolovov
Institute of Experimental Veterinary Science of Siberia and the Far East, Russian Academy of Agricultural Sciences
Russian Federation
T. M. Magerova
Veterinary Administration of Novosibirsk Region
Russian Federation
References
1. Лопатин С.В., Самоловов А.А. Ведущие факторы риска некробактериоза крупного рогатого скота // Ветеринария. - 2011. - № 5. - С. 9-11.
2. Горбань А.Н., Дунин-Барковский В.Л., Кирдин А.Н. и др. Нейроинформатика [Электронный ресурс]: http://neuroschool.narod.ru/books/,neurinf.html
3. Россиев Д.А., Головенкин С.Е. Прогнозирование осложнений инфаркта миокарда с помощью нейронных сетей // Нейроинформатика и ее приложения: тез. докл. II Всерос. Рабочего сем. (7-10 октября 1994 г.). - Красноярск, 1994. - С. 40.
4. Щетинин В.Г., Соломаха А.А. Применение искусственных нейронных сетей в клинической лабораторной диагностике [Электронный ресурс]: http://generation6.narod.ru/ med_app.htm
5. Царегородцев В.Г., Погребная Н.А. Нейросетевые методы обработки информации в задачах прогноза климатических характеристик и лесорастительных свойств ландшафтных зон [Электронный ресурс]: http: // neuroschool.narod.ru/pub/nn_klimat.html.
For citations:
Lopatin S.V.,
Samolovov A.A.,
Magerova T.M.
METHOD OF NEURONET EVALUATION OF INCIDENCE RATE FOR NECROBACILLOSIS IN CATTLE. Siberian Herald of Agricultural Science. 2014;(1):82-87.
(In Russ.)
Views:
157