MediLens: An AI-Powered Medicine Information and Assistance System

Abstract

Medication errors and difficulty accessing reliable
drug information remain significant challenges in modern healthcare.
MediLens is an AI-powered web-based medicine information
and assistance system designed to provide accurate,
accessible, and easy-to-understand medication details for users.
The system integrates Optical Character Recognition (OCR) to
identify medicines from images of drug labels or packaging and
retrieves verified information from trusted biomedical databases.
Natural Language Processing (NLP) and a cloud-based Large
Language Model (LLM) are used to generate contextual summaries
and answer user queries through an interactive conversational
assistant. The backend is implemented using the FastAPI
framework and communicates with external knowledge sources
and AI services through secure API integration. MediLens follows
a modular architecture that includes OCR processing, information
retrieval, and AI-driven response generation to ensure
scalability and reliability. Core functionalities include medicine
identification from images, structured drug information retrieval,
automated summarization, and interactive question answering.
By combining retrieval-based verification with conversational
AI, MediLens aims to improve public health literacy, reduce
misinformation about medications, and demonstrate the practical
application of artificial intelligence in healthcare information
systems.