scholarly journals Stabilised bias field: segmentation with intensity inhomogeneity

2016 ◽  
Vol 10 (4) ◽  
pp. 302-313 ◽  
Author(s):  
Jack Spencer ◽  
Ke Chen
Keyword(s):  
Convex Relaxation ◽  
Intensity Variation ◽  
Bias Field ◽  
Intensity Inhomogeneity ◽  
Variational Model ◽  
Great Success ◽  
Two Phase ◽  
Relaxation Methods ◽  
Piecewise Constant ◽  
Field Estimation

Automatic segmentation in the variational framework is a challenging task within the field of imaging sciences. Achieving robustness is a major problem, particularly for images with high levels of intensity inhomogeneity. The two-phase piecewise-constant case of the Mumford-Shah formulation is most suitable for images with simple and homogeneous features where the intensity variation is limited. However, it has been applied to many different types of synthetic and real images after some adjustments to the formulation. Recent work has incorporated bias field estimation to allow for intensity inhomogeneity, with great success in terms of segmentation quality. However, the framework and assumptions involved lead to inconsistencies in the method that can adversely affect results. In this paper we address the task of generalising the piecewise-constant formulation, to approximate minimisers of the original Mumford-Shah formulation. We first review existing methods for treating inhomogeneity, and demonstrate the inconsistencies with the bias field estimation framework. We propose a modified variational model to account for these problems by introducing an additional constraint, and detail how the exact minimiser can be approximated in the context of this new formulation. We extend this concept to selective segmentation with the introduction of a distance selection term. These models are minimised with convex relaxation methods, where the global minimiser can be found for a fixed fitting term. Finally, we present numerical results that demonstrate an improvement to existing methods in terms of reliability and parameter dependence, and results for selective segmentation in the case of intensity inhomogeneity.

scholarly journals A Multi-Scale Framework for Bias Field Estimation in MRI Brain Images

2018 ◽  
Vol 7 (4.10) ◽  
pp. 197
Author(s):  
Maryjo M George ◽  
Kalaivani S
Keyword(s):  
Intensity Variation ◽  
Bias Field ◽  
Intensity Inhomogeneity ◽  
Brain Images ◽  
Multi Scale ◽  
Field Estimation ◽  
Corrected Data ◽  
Quantitative Analyses ◽  
Mr Brain Images ◽  
Mr Brain

Intensity inhomogeneity is an artifact in MR brain images and causes intensity variation of same tissues on the basis of location of the tissue within the image. It is crucial to minimize this phenomenon to improve the accuracy of the computer-aided diagnosis. Unlike the several methods proposed in the past to minimize intensity inhomogeneity, this proposed method uses a pyramidal decomposition strategy to estimate the bias field in MR brain images. The bias field estimated from the proposed multi-scale framework can be effectively used for intensity inhomogeneity correction of the acquired MR data. The proposed methodology has been tested on simulated database and quantitative analyses in terms of coefficient of variation in grey matter and white matter tissue regions separately and combined coefficient of joint variation are assessed. The qualitative and quantitative analyses on the corrected data indicate that the method is effective for intensity inhomogeneity on brain MR images.  

INTERLEAVED EM SEGMENTATION FOR MR IMAGE WITH INTENSITY INHOMOGENEITY

2014 ◽  
Vol 26 (05) ◽  
pp. 1450058
Author(s):  
Jingjing Gao ◽  
Mei Xie ◽  
Yan Zhou
Keyword(s):  
Image Segmentation ◽  
Em Algorithm ◽  
Bias Field ◽  
Intensity Inhomogeneity ◽  
Mr Images ◽  
Tissue Segmentation ◽  
Mr Image ◽  
Em Method ◽  
Field Estimation ◽  
Mr Image Segmentation

Expectation–maximization (EM) algorithm has been extensively applied in brain MR image segmentation. However, the conventional EM method usually leads to severe misclassifications MR images with bias field, due to the significant intensity inhomogeneity. It limits the applications of the conventional EM method in MR image segmentation. In this paper, we proposed an interleaved EM method to perform tissue segmentation and bias field estimation. In the proposed method, the tissue segmentation is performed by the modified EM classification, and the bias field estimation is accomplished by an energy minimization. Moreover, the tissue segmentation and bias field estimation are performed in an interleaved process, and the two processes potentially benefit from each other during the iteration. A salient advantage of the proposed method is that it overcomes the misclassifications from the conventional EM classification for the MR images with bias field. Furthermore, the modified EM algorithm performs the soft segmentation in our method, which is more suitable for MR images than the hard segmentation achieved in Li et al.'s12 method. We have tested our method in the synthetic images with different levels of bias field and different noise, and compared with two baseline methods. Experimental results have demonstrated the effectiveness and advantages of the proposed algorithm.

scholarly journals Comparison of minimization methods for nonsmooth image segmentation

2018 ◽  
Vol 9 (1) ◽  
pp. 68-86 ◽  
Author(s):  
L. Antonelli ◽  
V. De Simone
Keyword(s):  
Optimization Problem ◽  
Convex Relaxation ◽  
Numerical Comparison ◽  
Two Phase ◽  
Piecewise Constant ◽  
First Order ◽  
Smoothing Methods ◽  
Minimization Methods ◽  
Basic Approaches ◽  
Nonsmooth Optimization Problem

Abstract Segmentation is a typical task in image processing having as main goal the partitioning of the image into multiple segments in order to simplify its interpretation and analysis. One of the more popular segmentation model, formulated by Chan-Vese, is the piecewise constant Mumford-Shah model restricted to the case of two-phase segmentation. We consider a convex relaxation formulation of the segmentation model, that can be regarded as a nonsmooth optimization problem, because the presence of the l1-term. Two basic approaches in optimization can be distinguished to deal with its non differentiability: the smoothing methods and the nonsmoothing methods. In this work, a numerical comparison of some first order methods belongs of both approaches are presented. The relationships among the different methods are shown, and accuracy and efficiency tests are also performed on several images.

Volume and surface coil simultaneous reception (VSSR) method for intensity inhomogeneity correction in MRI

2021 ◽  
pp. 1-12
Author(s):  
Lin Wu ◽  
Tian He ◽  
Jie Yu ◽  
Hang Liu ◽  
Shuang Zhang ◽  
...  
Keyword(s):  
Bias Field ◽  
Correction Method ◽  
Surface Coil ◽  
Intensity Inhomogeneity ◽  
Acquisition Time ◽  
Surface Coils ◽  
In Vivo Experiments ◽  
Inhomogeneity Correction ◽  
Volume Coil

BACKGROUND: Addressing intensity inhomogeneity is critical in magnetic resonance imaging (MRI) because associated errors can adversely affect post-processing and quantitative analysis of images (i.e., segmentation, registration, etc.), as well as the accuracy of clinical diagnosis. Although several prior methods have been proposed to eliminate or correct intensity inhomogeneity, some significant disadvantages have remained, including alteration of tissue contrast, poor reliability and robustness of algorithms, and prolonged acquisition time. OBJECTIVE: In this study, we propose an intensity inhomogeneity correction method based on volume and surface coils simultaneous reception (VSSR). METHODS: The VSSR method comprises of two major steps: 1) simultaneous image acquisition from both volume and surface coils and 2) denoising of volume coil images and polynomial surface fitting of bias field. Extensive in vivo experiments were performed considering various anatomical structures, acquisition sequences, imaging resolutions, and orientations. In terms of correction performance, the proposed VSSR method was comparatively evaluated against several popular methods, including multiplicative intrinsic component optimization and improved nonparametric nonuniform intensity normalization bias correction methods. RESULTS: Experimental results show that VSSR is more robust and reliable and does not require prolonged acquisition time with the volume coil. CONCLUSION: The VSSR may be considered suitable for general implementation.

Sign in / Sign up

Export Citation Format

Share Document