The work aims to construct a mapping relationship between the emotion of users in the neurology consulting room and the color matching design of the consulting room through BP neural network based on the theory of Kansei engineering. Firstly, emotional image words of the users in the neurology consulting room were obtained through literature review and interview, and the representative emotional image word pairs were selected by cluster analysis. Secondly, the case analysis method was used to analyze the spatial color matching settings of the neurology consulting room and extract typical color samples. The typical color matching design scheme was formed through matching, and a perceptual evaluation matrix was constructed through the semantic difference method. Finally, using Matlab software, a BP neural network model of space color matching design in the neurology consulting room was established. After verification, the model mapped the relationship between perceptual images and space color matching settings.This model was helpful to get the predictive value of perceptual image evaluation scores of color matching in the neurology consulting room, and guide the color design of the consulting room space. This method improves the accuracy of color design of consulting room space, and provides an important reference for color design of medical space.