<b>ОЦЕНКА КЛИНИКО-ФИЗИОЛОГИЧЕСКОГО СТАТУСА СОБАК ПРИ ИСПОЛЬЗОВАНИИ ПРЕЦИЗИОННЫХ РАЦИОНОВ, ОПТИМИЗИРОВАННЫХ МЕТОДАМИ МАШИННОГО ОБУЧЕНИЯ</b>
Optimization of diets for dogs of various age groups requires consideration of specific nutrient requirements at different stages of ontogenesis. The application of machine learning algorithms opens new opportunities for developing personalized feeding strategies. The study was conducted on 12 German Shepherd dogs divided into three age groups of four animals each: juveniles (4–9 months), adults (3–7 years), and geriatric (9–12 years). Experimental diets (ML-diets) were developed using multi-criteria optimization methods based on FEDIAF and NRC standards. Monitoring included clinical examinations, hematological and biochemical blood analysis, and coprological examination on days 0, 3, 7, 14, and 28 of the experiment. Transition to the ML-diet was characterized by a transitory leukocytic response on day 3, with subsequent stabilization of parameters by day 28. Glucose concentration reached physiological optimum in all groups. Liver enzyme activity decreased, and bilirubin normalized, particularly in the senior group. Nitrogenous metabolism parameters remained stable. The mineral profile was characterized by optimal calcium-to-phosphorus ratio in all groups. Coprological analysis revealed high digestibility of proteins and fats. No clinical signs of diet intolerance were detected. The obtained results confirm the safety and high biological value of ML-diets for dogs of various age groups. Pronounced positive changes in metabolic parameters, especially in senior animals, indicate the effectiveness of a personalized feeding approach. Machine learning methods enable creation of diets that meet age-specific requirements and promote maintenance of metabolic homeostasis. Further studies should be directed at validation of the developed diets in other dog breeds and expansion of age ranges. Long-term monitoring of the impact of ML-diets on quality of life and life expectancy in animals is recommended.
Сведения
- Опубликовано
- 2026 · Т. 82, № 3, С. 129-143
- Язык
- Русский
- ISSN
- 3134-8114 · 3134-8122
- Издатель
- West Kazakhstan Agrarian Technical University named after Zhangir Khan