AI applications in veterinary digital health: a systematic survey

Area of Science:

• Veterinary Digital Health
• Artificial Intelligence in Animal Health

Background:

• Veterinary medicine is undergoing a digital transformation powered by Artificial Intelligence (AI).
• AI applications span diagnostic imaging, disease prediction, clinical decision support, wearable monitoring, and telemedicine.

Purpose of the Study:

• To systematically survey and categorize AI applications in veterinary digital health.
• To evaluate the current state and taxonomy of AI and deep learning (DL) in veterinary medicine.
• To identify limitations and future research priorities for AI integration.

Main Methods:

• Systematic survey following the PRISMA 2020 framework.
• Inclusion of studies published between 2013 and 2025.
• Classification of 22 selected studies across major AI application domains.

Main Results:

• Significant advancements noted in automated radiography, cytology, disease detection (cardiovascular, dermatology, oncology), and clinical decision support.
• AI demonstrates potential to improve diagnostic accuracy, early disease detection, and clinical efficiency.
• Emerging Large Language Models (LLMs) show promise for clinical consultation but require further validation.

Conclusions:

• Widespread clinical adoption of AI is limited by fragmented datasets, species diversity, and lack of external/real-world validation.
• Rigorous validation is crucial for LLM-based veterinary applications before routine use.
• Standardized datasets, explainable AI (XAI), and interdisciplinary collaboration are essential for safe and ethical AI integration.