Magnetic Resonance Imaging (MRI) is a medical imaging modality that is commonly employed for the analysis of different diseases. However, these images come with several problems such as noise and other imaging artifacts added during acquisition process. The researchers have actual challenges for segmentation under the consideration of these effects. In medical images, a well-known clustering approach like Fuzzy C-Means widely used for segmentation. The performance of FCM algorithm is fast in noise-free images; however, this method did not consider the spatial context of the image due to which its performance suffers when images corrupted with noise and other imaging relics. In this paper, a weighted spatial Fuzzy C-Means (wsFCM) segmentation method is proposed that considered the spatial information of image. Moreover, a spatial function is also developed that integrate a membership function. In order assess this function, a neighborhood window is established around a pixel and more weights have been assigned to those pixels which have greater correlation with central pixel in local neighborhood. By integration of this spatial function in membership function, the modified membership function strengthens the original membership function in handling the noise and intensity inhomogeneity, which has the ability to preserves and maintains structural information like edges. A comprehensive set of experimentation is performed on publicly accessible simulated and real standard brain MRI datasets. The performance of the proposed method has been compared with existing state-of-the-art methods. The results show that the performance of the proposed method is better and robust in handling noise and intensity inhomogeneity than of the existing works.

磁共振成像（MRI）是一种医学成像方法，通常用于分析不同的疾病。但是，这些图像存在一些问题，例如在采集过程中添加了噪声和其他成像伪像。考虑到这些影响，研究人员在分割时面临着实际挑战。在医学图像中，众所周知的聚类方法（如模糊C均值）广泛用于分割。FCM算法在无噪声图像中的性能很快；但是，这种方法没有考虑图像的空间背景，因此当图像被噪声和其他成像文物破坏时，其性能会受到影响。本文提出一种考虑图像空间信息的加权空间模糊C均值（wsFCM）分割方法。此外，还开发了集成隶属度函数的空间函数。为了评估该功能，在像素周围建立邻域窗口，并且向与本地邻域中的中心像素具有更大相关性的那些像素分配了更多的权重。通过将此空间函数集成到隶属函数中，修改后的隶属函数在处理噪声和强度不均匀性方面加强了原始隶属函数，从而具有保留和维护边缘等结构信息的能力。在可公开访问的模拟和真实标准大脑MRI数据集上进行了全面的实验。所提出的方法的性能已与现有的最新技术进行了比较。