The segmentation algorithms
|
|
The segmentation algorithms |
|
|
J.-F. Mangin
|
| This paper outlines a fully automatic method for the correction of intensity nonuniformity in MR images. This method does not require any a priori model of the tissue classes. The basic idea is that entropy is a good measure of the image quality which can be minimized in order to overcome the bias problem. Therefore, the optimal correcting field is defined by the minimum of a functional combining the restored image entropy and a measure of the field smoothness. This measure stems from the usual physical analogy with membranes. A third term added to the functional prevents the optimal field from being uniformly null. The functional is minimized using a fast annealing schedule. The performance of the method is evaluated using both real and simulated data. |
BrainVISA doc
|
|
J.-F. Mangin, O. Coulon, and V. Frouin
|
| In this paper, we propose a robust fully non-supervised method dedicated to the segmentation of the brain in T1-weighted MR images. The first step consists in the analysis of the scale-space of the histogram first and second derivative. We show first that the crossings in scale-space of trajectories of extrema of different derivative orders follow regular topological properties. These properties allow us to design a new structural representation of a 1D signal. Then we propose an heuristics using this representation to infer statistics on grey and white matter grey level values from the histogram. These statistics are used by an improved morphological process combining two opening sizes to segment the brain. The method has been validated with 70 images coming from 3 different scanners and acquired with various MR sequences. |
BrainVISA doc
|
|
J.-F. Mangin, J. Régis, and V. Frouin
|
| This paper proposes an alternative to mathematical morphology to analyze complex shapes. This approach aims mainly at the detection of shape bottlenecks which are often of interest in medical imaging because of their anatomical meaning. The detection idea consists in simulating the steady state of an information transmission process between two parts of a complex object in order to highlight bottlenecks as areas of high information flow. This information transmission process is supposed to have a conservative flow which leads to the well-known Dirichlet-Neumann problem. This problem is solved using finite differences, over-relaxation and a raw to fine implementation. The method is applied to the detection of main bottlenecks of brain white matter network, namely corpus callosum, anterior commissure and brain stem. |
BrainVISA doc
|
|
J.-F. Mangin, V. Frouin, I. Bloch,
|
| We propose an algorithm allowing the construction of a structural representation of the cortical topography from a T1-weighted 3D MR image. This representation is an attributed relational graph (ARG) inferred from the 3D skeleton of the object made up of the union of gray matter and cerebro-spinal fluid enclosed in the brain hull. In order to increase the robustness of the skeletonization, topological and regularization constraints are included in the segmentation process using an original method: the homotopically deformable regions. This method is halfway between deformable contour and Markovian segmentation approaches. The 3D skeleton is segmented in simple surfaces (SSs) constituting the ARG nodes (mainly cortical folds). The ARG relations are of two types: first, the SS pairs connected in the skeleton; second, the SS pairs delimiting a gyrus. The described algorithm has been developed in the frame of a project aiming at the automatic detection and recognition of the main cortical sulci. Indeed, the ARG is a synthetic representation of all the information required by the sulcus identification. This project will contribute to the development of new methodologies for human brain functional mapping and neurosurgery operation planning. |
BrainVISA doc
|
|
J.-F. Mangin, F. Tupin, V. Frouin, I. Bloch,
|
| We propose to make low level segmentation of 3D medical images rely on deformable topological models. An initial label image made up of a set of deformable regions is endowed with a priori known topological properties of the desired result. Then, multi-scale topology preserving deformations are applied to this image in order to minimize a global energy whose form stems from the Gibbs Random Field domain. This new se gmentation framework leads to important improvements of previous works. First, topological regularization is included in the segmentation process which results in better robustness of posterior processing like skeletonization. Second, the topological arrangement allows discrimination between objects of similar intensities. Third, interface between two deformable regions behaves like a deformable surface but surface regularization relies on Ising model rather than on differential terms. This results in a better behaviour with highly convoluted objects like the cortex. Finally, this approach allows the management of interactions between several deformable objects. |
|
F. Poupon, J.-F. Mangin, D. Hasboun,
|
| We propose a new way of embedding shape distributions in a topological deformable template. These distributions rely on global shape descriptors corresponding to the 3D moment invariants. In opposition to usual Fourier-like descriptors, they can be updated during deformations at a relatively low cost. The moment-based distributions are included in a framework allowing the management of several simultaneously deforming objects. This framework is dedicated to the segmentation of brain deep nuclei in 3D MR images. The paper focuses on the learning of the shape distributions, on the initialization of the topological model and on the multi-resolution energy minimization process. Results are presented showing the segmentation of twelve brain deep structures. |