Artificial intelligence in veterinary diagnostic imaging: A literature review

Area of Science:

• Veterinary Medicine
• Artificial Intelligence
• Diagnostic Imaging

Background:

• Artificial intelligence (AI) is an emerging field in veterinary medicine.
• Machine learning (ML), a subset of AI, enables analysis of large imaging datasets for veterinary diagnostic imaging tasks.
• The current body of AI literature in veterinary imaging is small but growing.

Purpose of the Study:

• To review the existing literature on AI in veterinary diagnostic imaging.
• To provide background information for understanding AI applications in this field.
• To offer commentary on the current state and future potential of AI in veterinary imaging.

Main Methods:

• Literature review of peer-reviewed publications utilizing machine learning for imaging-associated tasks in veterinary clinical and biomedical research.
• Analysis of less than 40 identified publications across multiple anatomical regions.
• Author commentary based on the review findings.

Main Results:

• Fewer than 40 peer-reviewed publications currently exist on machine learning in veterinary imaging.
• Key challenges identified include data acquisition and quality, ground truth labeling, interdisciplinary collaboration, and the gap between academic research and commercial products.
• AI shows potential to enhance workflow, quality control, and image interpretation in veterinary radiology.

Conclusions:

• Despite challenges, AI, particularly machine learning, holds significant promise for advancing veterinary diagnostic imaging.
• Addressing data, collaboration, and commercialization hurdles is crucial for AI adoption.
• AI applications can help meet the increasing demand for radiological services in veterinary medicine.