Full automatic framework for segmentation of MR brain image

• Autores: Chong-Xun Zheng, Pan Lin, Yong Yang, Jian-Wen Gu
• Localización: Journal of Computer Science and Technology, ISSN-e 1666-6038, Vol. 5, Nº. 1, 2005 (Ejemplar dedicado a: Thirteenth Issue), págs. 6-11
• Idioma: inglés
• Enlaces
  • Texto completo
• Resumen

  • Magnetic Resonance Imaging is one of the most important medical imaging techniques for the investigating diseases of the human brain. A novel method for automatic segmentation Magnetic resonance brain image framework is proposed in this paper. This method consists of three-step segmentation procedures step. The method first uses level set method for the non-brain structures removal. Second, the bias correction method is based on computing estimates or tissue intensity distributions variation. Finally, we consider a statistical model method based on bayesian estimation, with prior Markov random filed models, for Magnetic resonance brain image classification. The algorithm consists of an energy function, based on the Potts model, which models the segmentation of an image. The algonthm was evaluated using simulated Magnetic resonance images and real Magnetic resonance brain images.