Multi-agent system for early prediction of urinary bladder inflammation disease.
ABSTRACT This paper presents an efficient real-time knowledge base architecture for multi-agent based patient diagnostic system for chronic disease management, basically, the early detection of Inflammation of urinary bladder and Nephritis of renal pelvis origin diseases. The model integrates information stored heterogeneous and geographically distributed healthcare centers. The paper presents two main contributions. First, a proposed multi-agent based system for mining frequent itemsets in distributed databases. Second, the implementation of this model on distributed medical databases in order to generate hidden medical rules. The proposed model can gather information from each department or from different hospitals, and using the cooperative agents it analyzes the data using association rules as a data mining technique. The proposed model improves the diagnostic knowledge and discovers the diseases based on the minimum number of effective tests, thus, providing accurate medical decisions based on cost effective treatments. It can also predict the existence or the absence of the diseases, thus improving the medical service for the patients. The proposed multi-agent system constitute an effort toward the design of intelligent, flexible, and integrated large-scale distributed data mining system.
Conference Paper: Fast Algorithms for Mining Association Rules in Large DatabasesProceedings of the 20th International Conference on Very Large Data Bases; 01/1994
Conference Paper: Agent Paradigm in Clinical Large-Scale Data Mining Environment.[Show abstract] [Hide abstract]
ABSTRACT: Intelligent agents are new paradigm for developing software applications. More than this, agent-based computing has been hailed as the next significant breakthrough in software development and the new revolution in classification techniques for large-scale data mining environment. Currently, agents are the focus of intense interest on the part of many sub-fields of computer science and artificial intelligence. In our work, we develop multi-agents platform gathering different type of agents. We provide to this platform the ability to operate automatically thanks to autonomous and intelligent agents. This new technology combines the agent approach with the monitoring strategy in order to automatically use the data mining for clinical analysis. Research into data mining in medical diagnosis is important to guide the clinicians in different phases of their diagnostic evaluations. The platform offers an interesting tool for data mining analysis using graph outputs and measures. An implementation of this platform on clinical database is presented. We discuss the importance of our approach and how it supports the data mining and also the possibility to generate and evaluate several association rules according to some scenarios predefined in the intelligent knowledge base of the proposed platform.Proceedings of the 2nd IEEE International Conference on Cognitive Informatics (ICCI 2003), 18-20 August 2003, London, UK; 01/2003
Conference Paper: Multi-agent systems: overview of a new paradigm for distributed systems[Show abstract] [Hide abstract]
ABSTRACT: Novel opportunities and challenges are emerging in the near future, such as Electronic commerce (e-commerce), due to the convergence of computers and telecommunication. Successful e-commerce requires rapid adaptation and excellent timing for service providers and users with guaranteed quality of service. Agent based technologies represent one of the most promising technological paradigm for distributed service system to meet these requirements. The aim of the current paper is to briefly review the agent systems basic architecture, with emphasize on agent communication languages. Two most popular agent communication languages, namely FIPA ACL and KQML have been briefly reviewed. The paper summarizes standardization activity of the agent system from the perspective of FIPA and OMG reference models. The relationship between FIPA and OMG has been given with a short discussion.High Assurance Systems Engineering, 2002. Proceedings. 7th IEEE International Symposium on; 02/2002