Validation of Tissue Modelization and Classification Techniques in T1-Weighted MR Brain Images

Abstract. We propose a deep study on tissue modelization and classification Techniques on T1-weighted MR images. Three approaches have been taken into account to perform this validation study. Two of them are based on Finite Gaussian Mixture (FGM) model. The first one consists only in pure Gaussian distributions (FGM-EM). The second one uses a different model for partial volume (PV) (FGM-GA). The third one is based on a Hidden Markov Random Field (HMRF) model. All methods have been tested on a Digital Brain Phantom image considered as the ground truth. Noise and intensity non-uniformities have been added to simulate real image conditions. Also the effect of an anisotropic filter is considered. Results demonstrate that methods relying in both intensity and spatial information are in general more robust to noise and inhomogeneities. However, in some cases there is no significant differences between all presented methods.

• Type: Conference
• Year: 2002
• Where: MICCAI
• Authors: Meritxell Bach Cuadra, Bram Platel, Eduardo Solanas, Torsten Butz, Jean-Philippe Thiran
• Keywords: Finite Gaussian Mixture, Intensity Non-uniformities, Medical Imaging, MICCAI 2002, Pure Gaussian Distributions