Abstract

In this paper, we present an agent-based system for distributed risk assessment of breast cancer development employing fuzzy and probabilistic computing. The proposed fuzzy multi agent system consists of multiple fuzzy agents that benefit from fuzzy set theory to demonstrate their soft information (linguistic information). Fuzzy risk assessment is quantified by two linguistic variables of high and low. Through fuzzy computations, the multi agent system computes the fuzzy probabilities of breast cancer development based on various risk factors. By such ranking of high risk and low risk fuzzy probabilities, the multi agent system (MAS) decides whether the risk of breast cancer development is high or low. This information is then fed into an insurance premium adjuster in order to provide preventive decision making as well as to make appropriate adjustment of insurance premium and risk. This final step of insurance analysis also provides a numeric measure to demonstrate the utility of the approach. Furthermore, actual data are gathered from two hospitals in Mashhad during 1 year. The results are then compared with a fuzzy distributed approach.

Links

Publisher Full Text

Authors

Tatari F, Akbarzadeh-T MR, Sabahi A

Institution

Department of Electrical Engineering, Center of Excellence on Soft Computing and Intelligent Information Processing, Ferdowsi University of Mashhad, Mashhad, Iran. fa_tatari@yahoo.com

Source

Journal of biomedical informatics 45:6 2012 Dec pg 1021-34

MeSH

Breast Neoplasms
Female
Fuzzy Logic
Humans
Insurance, Health
Probability
Risk Assessment

Pub Type(s)

Journal Article

Language

eng

PubMed ID

22692028