Brain Magnetic Resonance Image Segmentation Segmentation d’images IRM cérébrales

نویسندگان

  • Su Ruan
  • Stéphane Lebonvallet
چکیده

This paper presents two segmentation methods, based on Markov Random Fields (MRF), to quantify automatically brain tissue volumes from Magnetic Resonance (MR) images. Anatomically, a human brain is considered as composed of three main tissues: gray matter, white matter and cerebrospinal fluid (CSF). Due to the resolution limitation of the image acquisition system, a voxel in MR image may be composed of a mixture of several tissue types. The voxel is called mixel in this case and the phenomenon is called partial volume effects. In order to accurately measure each tissue volume, we consider that in a brain image volume there are not only the three main types of brain tissue, called pure classes in term of image processing, but also mixtures of the brain tissues called mixclasses. Taking into account the problem of partial volume effects, we propos the first method which uses two steps: 1) segmentation of the brain into pure and mixclasses using a mixture model, 2) reclassification of the mixclasses into the pure classes using the knowledge about them obtained from the first step. Both steps use MRF models. In order to obtain a sub-voxel resolution, we propose the second method which aims at calculating fuzzy memberships in each voxel to indicate the partial volume degree. A fuzzy MRF is proposed to model the partial volume degree. The both proposed methods are performed and evaluated on simulated and real MR images. RÉSUMÉ. Cet article présente deux méthodes de segmentation, basées sur les champs aléatoires de Markov (MRF), afin de quantifier automatiquement les volumes des tissus cérébraux à partir d’images IRM. Anatomiquement, un cerveau humain est considéré comme la composition de trois principaux tissus: la matière grise, la matière blanche et le liquide Studia Informatica Universalis. IRM segmentation 129 céphalo rachidien. Due à la limite de la résolution du système d’acquisition d’images, un voxel dans une image IRM peut tre considéré comme un mélange de plusieurs types de tissus. Dans ce cas, le voxel est appelé mixel, et le phénomène est appelé effet de volume partiel. Afin de mesurer précisément chaque volume de tissu, nous considérons que dans un volume cérébral, il y a non seulement les trois principaux tissus, nommées classes pures en terme de traitement d’image, mais aussi le mélange des tissus appelé mixclasses. En considérant le problème de l’effet de volume partiel, nous proposons une première méthode qui utilise deux étapes : 1) segmentation d’un cerveau en classes pures en utilisant un modele de mixture, 2) reclassification des mixclasses en trois classes pures à partir des connaissances obtenues lors de la première étape. Les deux étapes utilisent les modèles MRF. Afin d’obtenir une résolution sous-voxel, nous proposons une seconde méthode qui permet de calculer les appartenances dans chaque voxel pour indiquer les degrés du volume partiel. Les champs de Markov flous sont proposés pour modéliser le phénomène de l’effet de volume partiel. Les deux méthodes proposées sont testées et évaluées sur les images IRM simulées et réelles.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Improving Brain Magnetic Resonance Image (MRI) Segmentation via a Novel Algorithm based on Genetic and Regional Growth

Background: Regarding the importance of right diagnosis in medical applications, various methods have been exploited for processing medical images solar. The method of segmentation is used to analyze anal to miscall structures in medical imaging.Objective: This study describes a new method for brain Magnetic Resonance Image (MRI) segmentation via a novel algorithm based on genetic and regiona...

متن کامل

Segmentation of Magnetic Resonance Brain Imaging Based on Graph Theory

Introduction: Segmentation of brain images especially from magnetic resonance imaging (MRI) is an essential requirement in medical imaging since the tissues, edges, and boundaries between them are ambiguous and difficult to detect, due to the proximity of the brightness levels of the images. Material and Methods: In this paper, the graph-base...

متن کامل

Quantitative Comparison of SPM, FSL, and Brainsuite for Brain MR Image Segmentation

Background: Accurate brain tissue segmentation from magnetic resonance (MR) images is an important step in analysis of cerebral images. There are software packages which are used for brain segmentation. These packages usually contain a set of skull stripping, intensity non-uniformity (bias) correction and segmentation routines. Thus, assessment of the quality of the segmented gray matter (GM), ...

متن کامل

A Novel Fuzzy-C Means Image Segmentation Model for MRI Brain Tumor Diagnosis

Accurate segmentation of brain tumor plays a key role in the diagnosis of brain tumor. Preset and precise diagnosis of Magnetic Resonance Imaging (MRI) brain tumor is enormously significant for medical analysis. During the last years many methods have been proposed. In this research, a novel fuzzy approach has been proposed to classify a given MRI brain image as normal or cancer label and the i...

متن کامل

A hierarchical Convolutional Neural Network for Segmentation of Stroke Lesion in 3D Brain MRI

Introduction: Brain tumors such as glioma are among the most aggressive lesions, which result in a very short life expectancy in patients. Image segmentation is highly essential in medical image analysis with applications, particularly in clinical practices to treat brain tumors. Accurate segmentation of magnetic resonance data is crucial for diagnostic purposes, planning surgical treatments, a...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2010