Artificial Intelligence-Assisted Selection Strategies in Sheep: Linking Reproductive Traits with Behavioral...

Area of Science:

• Animal Science
• Agricultural Technology
• Artificial Intelligence in Agriculture

Background:

• Sheep farming productivity relies heavily on reproductive efficiency.
• Traditional methods for assessing reproductive traits are subjective and provide delayed feedback.
• Advancements in AI offer novel approaches for monitoring animal behavior.

Purpose of the Study:

• To review AI-powered behavioral monitoring tools for sheep reproduction.
• To propose a conceptual model (ReproBehaviorNet) linking behaviors to AI applications.
• To support real-time decision-making in sheep farming systems.

Main Methods:

• Utilizing AI, including video tracking, wearable sensors (accelerometers, GPS), and machine learning algorithms.
• Non-invasive, continuous monitoring of sheep behavior.
• Developing computer vision models for data analysis.

Main Results:

• AI enables identification of behavior-based indicators for estrus, lambing, and maternal behavior.
• Integration of sensors and AI leads to earlier detection of reproductive events and improved breeding precision.
• ReproBehaviorNet model maps behaviors to biological processes and AI tools.

Conclusions:

• AI-assisted behavioral phenotyping can enhance genetic progress, animal welfare, and sustainability in sheep farming.
• Challenges include data quality, infrastructure costs, and technical expertise requirements.
• Interdisciplinary collaboration is crucial for effective and equitable adoption of AI technologies.