VetClaw: An Edge-Cloud Multimodal Agentic System for Veterinary Disease Screening
Presents cutting-edge research on multimodal agentic systems with practical veterinary applications and architectural innovations.
AI Summary
Researchers present VetClaw, an edge-cloud agentic system for veterinary disease screening that combines edge sensing with server-hosted vision-language models. The system uses OpenClaw for edge orchestration and LangGraph for workflow management, demonstrating improved zero-shot classification performance with multimodal inputs compared to image-only approaches. This represents a novel application of agentic systems in veterinary medicine with safety-aware workflow coordination.
Excerpt
We present VetClaw, an edge-cloud multimodal agentic system for early veterinary disease screening. VetClaw uses a camera module as an edge sensing device and sends captured images, together with optional symptom descriptions, to a server-hosted vision-language model for zero-shot disease classification. The system separates agent interaction from workflow orchestration: OpenClaw provides scheduling, tool access, user interaction, and notification services on the edge device, while LangGraph man