Abstract

Leveraging artificial intelligence (AI) approaches in animal health (AH) makes it possible to address highly complex issues such as those encountered in quantitative and predictive epidemiology, animal/human precision-based medicine, and the study of host–pathogen interactions. This paper presents the design, development, and evaluation of an AI-Powered PetVet Assistant – a comprehensive platform for smart disease detection and personalised care recommendations for pets. The system integrates deep learning-based image classification, Retrieval-Augmented Generation (RAG), Natural Language Processing (NLP), and a real-time veterinary appointment booking module. A two-phase fine-tuned MobileNetV2 model achieves a validation accuracy of approximately 73.8% on the Animal Skin Disease dataset, outperforming a baseline CNN trained from scratch by 25.1 percentage points. The combined image and RAG pipeline satisfies a sub-5-second end-to-end latency target, demonstrating feasibility for practical veterinary deployment.

Authors

Atul Singh, Danish Ansari, Farooqui Luckman, Shaikh Mohammed Hashim, Sachin Charbe
Rizvi College of Engineering, India

Keywords

Animal Health, Pet Disease Detection, Deep Learning, MobileNetV2, Retrieval-Augmented Generation