Lab4U has launched an AI agent service that allows users to generate a laboratory test order by uploading a doctor's prescription or describing their health goals in natural language. The system automatically identifies relevant tests from the company's catalog, verifies availability in the user's city, and creates a pre-filled shopping cart.
The service utilizes the Model Context Protocol (MCP) to connect AI models directly to the laboratory's database. This integration enables the AI to access real-time information on 780 available tests and medical complexes across 144 cities, moving beyond simple text-based recommendations to actionable order fulfillment.
During internal testing with OpenAI and Anthropic models, the agent successfully formed a shopping cart in 92% of 132 scenarios, with a 90% accuracy rate compared to expert recommendations. The company does not share patient medical history or personal data with the AI, as user details are only entered during the final checkout on the Lab4U website.
Lab4U intends to expand the service by integrating it with corporate AI agents from clinics, insurance companies, and telemedicine providers. Future updates are expected to include features for locating the nearest medical center, scheduling appointments, tracking order status, and facilitating repeat orders.
