关键词: t-混合模型, 核磁共振成像（MRI）分割, 白质

Abstract: As Magnetic Resonance Imaging（MRI） is an important technology of clinical diagnosis and research，the accuracy of the MR image segmentation directly influences the validity of following processing.The segmentation method based on t-mixture model receives attention because of its speediness and robustness among current methods.The main procedure is outlined as follows.Firstly，the parameters of t-mixture model are estimated，and then the posterior probability of the pixels of the image is computed.At last the image is segmented according to Bayes decision rule for minimum error.By analyzing the features of MR images，the algorithm of white matter segmentation of brain MR images based on t-mixture model is proposed，and better experimental results are obtained.

Key words: t-mixture model, Magnetic Resonance Imaging（MRI） segmentation, white matter