Applications of Artificial Intelligence in Veterinary Ophthalmology: A Comprehensive Review of Recent Advances

Background: Artificial intelligence (AI) has become an emerging tool in veterinary ophthalmology, helping veterinarians detect ocular diseases at early stages with higher accuracy. Deep learning (DL) algorithms can automatically analyze medical images, reducing dependence on subjective human evaluation. Methods: This review summarizes twelve key studies that applied AI models, including Convolutional Neural Networks (CNNs), YOLOv۵, DenseNet, and Vision Transformers (ViT), to diagnose ocular diseases such as cataracts, corneal ulcers, uveitis, and dry eye disease (DED) in different animal species. Each study was assessed based on model type, dataset, diagnostic task, and performance. Results: CNN and DenseNet models achieved high accuracy in disease classification and segmentation, YOLOv۵ provided real-time detection of tear film disorders, and ViT enhanced recognition of complex ocular features. Despite these promising results, challenges remain due to small datasets, lack of standardization, and limited interpretability of deep models. Conclusion: AI-based systems demonstrate strong potential to improve diagnostic precision, speed, and accessibility in veterinary ophthalmology. Future research should focus on developing interpretable, multimodal, and portable AI models trained on large, multi-institutional datasets to better support clinical decision-making.

کلیدواژه ها:

Artificial Intelligence ، Veterinary Ophthalmology ، Animal Disease Diagnosis ، Deep Learning ، Eye Disease