Context: Nowadays, functional MRI is widely used in the study of brain function. If it has the advantage of being non-invasive it has several deficiencies. Zones that activate in response to a motor stimulus mislead the analyst in view of the complexity and differences between individuals and their intrinsic specificities. In the case of a brain tumor located near these areas, surgery will not be without risk on the motricity of the organ in question.
Methods: In this study, a fuzzy logic analysis is proposed. In view of the complexity of the system, the variables that define the fMRI images are considered as imprecise variables and thus as fuzzy variables. The motor stimulus, effect of the tumor and parasitic movements are considered as inputs to the system. The quality of the resulting image expresses the output variable. The system done, allows adjusting the input variables for an optimal image.
Results: The input or output variables are fuzzyfied. This operation makes it possible to pass numerical values to linguistic variables in the image of human reasoning. The uncertainties are compensated. The basis of the rules established from the real cases makes it possible to reproduce faithfully the state of the patient. The result at the output is a function of the input variables with the support of all possible combinations.
Conclusions: The proposed system allows locating the motor area with precision preoperatively. When the tumor affects the targeted motor zone, it will be possible to intervene with the slightest risk.