Fuzzy c-means algorithm and its variants are popular for the segmentation of magnetic resonance imaging (MRI) data. The enhanced fuzzy c-means approach is one among them that comprises weighted local spatial data. However, the quantity of spatial data added with input MRI image differs and that depends on the noise content and sequence of MRI. Hence, the value of weight factor needs to be chosen appropriately and automatically to attain the accurate segmentation results. In this perspective, the current work focuses to generate optimum weight values and presents an optimized enhanced fuzzy c-means algorithm for MRI data. The proposed method utilizes the multi-objective particle swarm optimization to control the weight parameter that leads to maximum segmentation accuracy. The new approach is tested and validated on a standard simulated BrainWeb MRI dataset. The outcome shows that the proposed approach is flexible and robust to noise content as compared to the conventional algorithms.