The recognition of migraine headache by designation of fuzzy expert system and usage of LFE learning algorithm

Abstract

The migraine headache is a kind of most populated headache which its rate of population is so high. The first step for starting of treatment is the recognition stage. Also the fuzzy logic has good power for describing of enigmatic and imprecise aspects and due to this reason this tool could be used for the system modeling. The aim of this research is the migraine recognition by the usage of fuzzy logic and systems. A fuzzy expert system for diagnosis of migraine by LFE algorithm is presented, that Mamdani model was used in fuzzy inference engine using MAX-MIN as OR-AND operators and Centroid method was used as defuzzification technique. By the usage of 148 patients, the migraine diagnostic system has been trained by LFE algorithm and in average 80 pieces of IF-THEN rules have been produced for fuzzy system and accuracy, precision, sensitivity, specificity of the system were 97%, 80%, 70%, 94%. By attention to this point that the linguistic rules may be incomplete when human expert to express their knowledge and according to importance of early diagnosis and favorable results, the LFE training algorithm rather than human experts system, will be more effective for recognition of migraine headache. © 2017 IEEE.