A Fuzzy Rule-Based Expert System for Diagnosing Asthma

Asthma is a chronic lung disorder of which the number of su erers estimated to be between 1.4-27.1% of the population in di erent areas of the world. Results of various studies show that asthma is usually under-diagnosed, especially in developing countries, because of limited access to medical specialist and laboratory data. The purpose of this paper is to design a fuzzy rule-based expert system to alleviate this hazard by diagnosing asthma at initial stages. A knowledge representation of this system is provided from a high level, based on patient perception, and organized into two di erent structures called Type A and Type B. Type A is composed of six modules, including symptoms, allergic rhinitis, genetic factors, symptom hyper-responsiveness, medical factors and environmental factors. Type B is composed of 8 modules including symptoms, allergic rhinitis, genetic factors, response to short-term drug use, bronchodilator tests, challenge tests, PEF tests and exhaled nitric oxide. The nal result of every system is de-fuzzi ed in order to provide the assessment of the possibility of asthma for the patient. Veri cation and validations criteria are considered throughout a life-cyclethe system was developed by the participation of general physicians, experienced asthma physicians and asthmatic patients.