A Review Paper On An Integrated Multi-Agent Decision Support System for Medical Image Analysis, Clinical Diagnosis, Drug Interaction Assessment, and Risk Evaluation

Healthcare decision making often requires multiple doctors to consider different types of information like patient symptoms, their medication history, medical images, lab reports, and relevant medical literature. All this information is frequently fragmented across multiple sources. A multi-agent decision support system can help to integrate all these sources and organize them. This paper presents the design and development of a Multi-Agent Healthcare Decision Support System (DSS) designed to integrate medical image analysis, clinical diagnosis, drug-interaction assessment, evidence retrieval, and risk evaluation through a modular architecture. The proposed system consists of 6 different specialized agents namely vision agent, diagnosis agent, drug-interaction agent, literature/RAG agent, emergency risk agent and Coordinator/Orchestrator. A single agent is not able to do all tasks. Image analysis, medication assessment, evidence retrieval, and risk evaluation have different computational requirements and need different validation procedures for each agent. The system’s effectiveness must be established through systematically designing, developing and testing the individual agent and integrating the whole workflow. The proposed system aims to support clinical decision making, not to replace professionals.