View Proposal
-
Proposer
-
Radu-Casian Mihailescu
-
Title
-
Multi-Agent Argumentation Frameworks with LLM-Augmented Reasoning in Healthcare Systems
-
Goal
-
-
Description
- The project explores multi-agent systems (MAS) combined with large language models (LLMs) to support clinical decision-making through structured argumentation. Each agent represents a distinct medical perspective (e.g., conservative vs. risk-tolerant) and debates treatment options using symbolic reasoning frameworks, while LLMs express arguments in fluent, clinician-style language.
The system aims to improve explainability and trust in AI by making reasoning processes transparent, interactive, and aligned with real-world clinical workflows. A prototype will integrate symbolic reasoning, retrieval-augmented LLMs, and intuitive interfaces, enabling clinicians to follow, contest, and interact with AI-generated justifications. This human-in-the-loop design fosters human-centered AI by combining machine reasoning with medical expertise, addressing challenges in trust, explainability, and low-data adaptation.
- Resources
-
-
Background
-
-
Url
-
-
Difficulty Level
-
Moderate
-
Ethical Approval
-
None
-
Number Of Students
-
2
-
Supervisor
-
Radu-Casian Mihailescu
-
Keywords
-
-
Degrees
-
Bachelor of Science in Computer Science
Bachelor of Science in Computer Systems
Bachelor of Science in Information Systems
Bachelor of Science in Software Development for Business (GA)
Bachelor of Science in Computing Science
Bachelor of Engineering in Robotics
Bachelor of Science in Computer Science (Cyber Security)
Bachelor of Science in Statistical Data Science
BSc Data Sciences