Dirichlet Markov Random Field Segmentation Of Brain MR Images

摘要：

Accurate segmentation of magnetic resonance imagesaccording to tissue type is widely studied by many researchers,Recently Markov Random Field (MRF) has been used in thisarea. However the original MRF is supervised. So we introduce anovel approach called Dirichlet Markov Random Field forMagnetic Resonance Image (MRI) brain tissue classification. Theapproach uses Dirchilet Process Mixture (DPM) to get localinformation of MRFs energy function. Instead of finitecomponent, DPM use infinite component, in which the priordistribution is defined on the space of all possible distribution.But efficient implementations of the DP mixture model can beslowly to converge and their convergence can be difficult todiagnose with the Markov Chain Monte Carlo (MCMC) methodsfor sampling from the posterior distribution of the parameters.So this algorithm uses variational Bayesian (VB) approximationsto the DP mixture model. Experiment result proves thisalgorithm can segment the MRI smoothly and accurately.

作者: Wentao Wang Cong Chen

作者单位: College of Computer Science,South-Central University for Nationalities Wuhan , 430074,China

会议类型: 国际会议

会议名称: The 4th International Conference on Bioinformatics and Biomedical Engineering(第四届IEEE生物信息与生物医学工程国际会议 iCBBE 2010)

会议地点: 成都

会议语种:英文

页码: 1-4

在线出版日期: 2010-06-18（万方平台首次上网日期，不代表论文的发表时间）