Dairy farming sustainability A systematic review to understand mastitis and lameness detection using machine learning (ML) /
Mastitis and lameness remain major economic issues in dairy farming, worsened by climate change. Automation and machine learning are increasingly used in cattle health management; nevertheless, current reviews predominantly discuss precision livestock farming in general health management, lacking a...
Elmentve itt :
| Szerzők: | |
|---|---|
| Dokumentumtípus: | Cikk |
| Megjelent: |
2026
|
| Sorozat: | ANNALS OF AGRICULTURAL SCIENCES (CAIRO)
71 No. 1 |
| Tárgyszavak: | |
| doi: | 10.1016/j.aoas.2026.100412 |
| mtmt: | 37107254 |
| Online Access: | http://publicatio.bibl.u-szeged.hu/40418 |
| Tartalmi kivonat: | Mastitis and lameness remain major economic issues in dairy farming, worsened by climate change. Automation and machine learning are increasingly used in cattle health management; nevertheless, current reviews predominantly discuss precision livestock farming in general health management, lacking a specific comparison of machine-learning methodologies for mastitis and lameness detection. We reviewed 64 articles on machine learning applications in cattle health, focusing on udder and locomotor diseases. Our review shows a significant increase in the use of machine learning approaches to address these health problems since 2018 onward, reflecting a broader shift toward data-driven strategies in cattle health management. Literature analyzes various data types to predict diseases, identify pathogens, and develop predictive models. Because different algorithms are better suited to specific data types and research goals, and studies vary widely in data sources, case definitions, and evaluation protocols, the current evidence does not support the context-independent superiority of any single method. However, reliable machine learning methods for cattle health management still face significant challenges with labeled data scarcity, imbalanced datasets and lack of standardized data collection. This approach has the potential to reduce human effort and the time required for livestock health monitoring. However, further research is needed to better understand this technology and enhance its applicability under various farm conditions. |
|---|---|
| Terjedelem/Fizikai jellemzők: | 10 |
| ISSN: | 0570-1783 |