Noha Youssry El-Zehiry

A graph cut based active contour for multiphase image segmentation

In this paper, we introduce a novel hierarchical approach for multiphase image segmentation. The approach presents a unified framework that unifies two basic segmentation approaches; level set methods and graph cut algorithms. In the work of El-Zehiry et al. (2007), we have presented a bimodal image segmentation approach that have the advantages of the level set methods such as robustness to noise, blurred edges and topology changes and the advantages of graph cuts vis-a-vis global optimization and speed. Our main objective in this paper is to extend our previous approach to segment n classes. The results section will show that our algorithm outperforms the multiphase image segmentation approach introduced in the work of Vese and Chan (2002).

A graph cut based active contour without edges with relaxed homogeneity constraint

The paper presents a graph cut based active contour without edges segmentation model to track pedestrian in thermal images. The deformable model is based on the Mumford- Shah piecewise constant energy formulation. However, the model presented here relaxes the global homogeneity assumption of the Mumford- Shah functional. A discrete energy formulation is presented and the optimization is performed using graph cuts. The major advantages of our approach are; 1) The optimization using graph cuts makes the segmentation process much faster than solving it using level sets. 2) Relaxing the global homogeneity assumption makes the model more practical.

Graph cut based deformable model with statistical shape priors

This paper presents a novel graph cut based segmentation approach with shape priors. The model incorporates statistical shape prior information with the active contour without edges model . Our model also relaxes the homogeneity constraint that assumes that the image is modeled by a piecewise constant approximation. The major contribution of this paper is to present a graph cut optimization for the energy function. Hence, the resultant approach is a fully automatic shape based segmentation approach that is insensitive to initialization and does not require any user interaction. Due to the polynomial time complexity of graph cut optimization approaches, our segmentation technique is much faster than the state of the art deformable models segmentation approaches.

Variability of the relative corpus callosum cross sectional area between dyslexic and normally developed brains

Minicolumnar disturbance is a common feature to both dyslexic and autistic brains. This paper is motivated by the persistent need to investigate the effect of minicolumnar disturbance on the magnetic resonance images of the brain. This will serve as a preliminary step to develop a non-invasive methodology to discriminate between the diseases based on the MRI findings. In this paper, we investigate the variability of the ratio between the corpus callosum cross sectional area and the total brain intracranial volume between two groups; a group of dyslexic patients and another group of normal controls. The results show that this ratio differs significantly between the two groups and that it can be used as a discriminatory feature between dyslexic brains and typically developed ones.

Effect of minicolumnar disturbance on dyslexic brains: an MRI study

The minicolumn is generally considered the basic unit of the neocortex in all the mammalian brains. Enlargement of the cortical surface is believed to occur through the addition of minicolumns rather than a single neuron. This study aims at testing the hypothesis that brain developmental disorders can be diagnosed and analyzed in terms of the minicolumnar disturbance. To do this, we propose to correlate the pathological findings in terms of the minicolumnar structure to the MRI findings in terms of volumetric analysis

An active surface model for volumetric image segmentation

In this paper, we introduce a graph cut based active surface model that incorporates graph cuts optimization tools with implicit surface representation to solve the segmentation problem. We will introduce a discrete formulation of the surface evolution problem, prove that the discrete energy function is graph representable and propose how to optimize it using graph cuts. The advantage of this model is two fold: First, Graph cuts are mostly global optimization tools which makes the model very robust and not sensitive to initialization, moreover, the dynamic labeling associated with graph cuts optimization tools makes the model very fast. Second, the implicit representation of the surface makes it robust to topology changes.

Pathological vocal folds diagnosis using modified active contour models

This paper presents the study of vocal videostroboscopic videos to detect morphological pathologies using an active contour segmentation and objective measurements. The ad-hoc designed active contour algorithm permits to obtain a robust and fast segmentation using vocal folds images in RGB format. In this work, we have employed connected component analysis as a post-processing tool. After the segmentation process the image is analyzed and the pathology can be localized automatically and we can extract some features of each fold (such as the size of the polyp or cyst, the glottal space, the position…).

MRI brain extraction using a graph cut based active contour model

Brain extraction refers to stripping the skull and removing any non brain tissue such as fat, bone and eye balls from the MRI of the head. Brain extraction is an extremely important preliminary step before any brain analysis algorithm. This paper proposes a novel algorithm for the extracting the brain tissue using a graph cuts based active contour model. The model combines the implicit curve evolution techniques with graph cuts optimization tools to provide a fast and robust segmentation algorithm. A discrete version of the Mumford Shah functional will be presented and the optimization will be performed on a discrete lattice using the max-flow/min-cut algorithm. The implicit curve evolution is performed by iteratively minimizing the discrete function and is simply described as follows: we will construct a graph in which each pixel in the image has a corresponding vertex and we will add two auxiliary vertices (Source (S) and Target (T)) that will later represent the labeling and this will complete the vertex set of the graph. The edge set of the graph will consists of two subsets: Terminal links that connect each vertex to either the source or the target, the weights of these links represent the external energy of the active contour model and according to the Mumford-Shah functional will be calculated as the intensity deviation of the corresponding pixel around the mean value of the class of interest. Neighboring Links: these will connect the neighboring vertices with each other and, the weights of these links will represent the internal energy of the active contour and are calculated from the length of the evolving contour. A discrete approximation of the Euclidean length of the contour is presented using the Cauchy Crofton formula. After constructing the graph, a max-flow/min-cut algorithm will be applied to find the minimum cut. The minimum cut will subdivide the vertices of the class into two disjoint sets one of them contains the source and the other contains the sink, respectively. The pixels that correspond to all the vertices in the first set will be have a label 1 and all the other pixels will have a label zero and this terminates the labeling process. Applying the curve evolution model on the MRI slice, it will tend to group the more homogeneous tissue in one class and hence the white matter and gray matter tissues will be grouped with the fat and everything else in the other class. Fat is naturally far apart from the gray matter and white matter and hence, the curve evolution algorithm is followed by a connected component analysis that picks the most dominant component/s as the brain tissue. The advantages of our algorithm over the currently existing brain extraction algorithms are summarized as follows: 1) Graph cuts are considered as a global optimization tool and hence our model is less prone to error and not sensitive to initialization. 2) Graph cuts can obtain the global minimum of most functions in polynomial time, which makes our algorithm very fast when compared to most of the brain extraction techniques that mostly depend on level sets implementations. 3) The implicit curve representation makes the model very robust to topology changes.