Case-Based-Reasoning System for Feature Selection and Diagnosing Disease; Case Study: Asthma

Abstract

Asthma is a chronic informatory disease of the respiratory canals in which it has not become obvious what is the reason for the reports argumentation on the ground of asthma prevalence. In the present research, the purpose would be to design a case-based-reasoning (CBR) model in order to assist a physician to diagnose the type of disease and also the needed therapy.

At first for designing this system, the disease variables were discriminated and were at the patients' disposal as a questionnaire, and after gathering the relevant data (CBR) algorithm was rendered on the data which led to the asthma diagnosis. The system was tested on 325 asthmatic and non asthmatic adult cases and was accessed with eighty percent accuracy. The consequences were promising. With regard to the fact that the factors of the disease are different in various countries, This study was performed in order to determine risk factors for asthma in Iranian society and the results of research showed that the most important variables of asthma disease in Iran are symptoms heperresponsivity, frequency of cough, cough.

Key words: data mining, case based reasoning, asthma, diagnosis.