Medicine Assistant
An intelligent assistant for diabetes care — centralize patient data, track key metrics, and generate guideline-aware recommendations.
Why it exists
To reduce cognitive load for clinicians by surfacing relevant patient history, trends, and evidence-based suggestions during consultations.
Who it's for
Primary care providers, endocrinologists, diabetes educators, and clinical researchers managing diabetes populations.
Trust & Safety
Outputs are suggestions only — always reviewed by a clinician. Data privacy and secure storage are priorities for deployments.
Our Approach
We leverage Large Language Models (LLMs) together with structured clinical data to provide concise, actionable recommendations. The core component — MedicineAssistantAgent — synthesizes recent labs, vitals, and medications to highlight opportunities for optimization.
The agent is a decision support tool, designed to augment clinical judgment, not replace it. All recommendations should be validated by a qualified healthcare professional before changes are applied.
Key Features
- Patient directory with fast search and list/grid views.
- Consultation assistant that ingests latest labs and symptoms.
- AI-generated recommendations aligned to clinical guidelines.
- Exportable reports and notes for clinical documentation.
Technology & Integrations
Built with a Python backend and a lightweight frontend. Optionally integrates with LLM providers for natural language reasoning and can be connected to EHR systems for data exchange.
Stack: Flask, Tailwind CSS, ChromaDB (local), optional LLM providers.
Team & Contribution
This project was created by the Zakey-Team-1. Contributions are welcome — please open issues or pull requests on the repository to suggest improvements or report bugs.