A Dirichlet process mixture model for brain MRI tissue classification

• Published in: Medical image analysis Vol. 11; no. 2; pp. 169 - 182
• Main Author: Ferreira da Silva, Adelino R.
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier B.V 01.04.2007
• Subjects: Automatic brain MRI classification; Bayes Theorem; Brain - anatomy & histology; Dirichlet process mixture model; Humans; Image Processing, Computer-Assisted; Magnetic Resonance Imaging; Models, Theoretical; Nonparametric Bayesian estimation; Robust estimation; Automatic brain MRI classification; Dirichlet process mixture model; Robust estimation; Magnetic resonance imaging; Nonparametric Bayesian estimation
• ISSN: 1361-8415; 1361-8423
• DOI: 10.1016/j.media.2006.12.002

Accurate classification of magnetic resonance images according to tissue type or region of interest has become a critical requirement in diagnosis, treatment planning, and cognitive neuroscience. Several authors have shown that finite mixture models give excellent results in the automated segmentation of MR images of the human normal brain. However, performance and robustness of finite mixture models deteriorate when the models have to deal with a variety of anatomical structures. In this paper, we propose a nonparametric Bayesian model for tissue classification of MR images of the brain. The model, known as Dirichlet process mixture model, uses Dirichlet process priors to overcome the limitations of current parametric finite mixture models. To validate the accuracy and robustness of our method we present the results of experiments carried out on simulated MR brain scans, as well as on real MR image data. The results are compared with similar results from other well-known MRI segmentation methods.