Superpixel and multi-atlas based fusion entropic model for the segmentation of X-ray images.

X-ray image segmentation is an important and crucial step for three-dimensional (3D) bone reconstruction whose final goal remains to increase effectiveness of computer-aided diagnosis, surgery and treatment plannings. However, this segmentation task is rather challenging, particularly when dealing with complicated human structures in the lower limb such as the patella, talus and pelvis. In this work, we present a multi-atlas fusion framework for the automatic segmentation of these complex bone regions from a single X-ray view. The first originality of the proposed approach lies in the use of a (training) dataset of co-registered/pre-segmented X-ray images of these aforementioned bone regions (or multi-atlas) to estimate a collection of superpixels allowing us to take into account all the nonlinear and local variability of bone regions existing in the training dataset and also to simplify the superpixel map pruning process related to our strategy. The second originality is to introduce a novel label propagation step based on the entropy concept for refining the resulting segmentation map into the most likely internal regions to the final consensus segmentation. In this framework, a leave-one-out cross-validation process was performed on 31 manually segmented radiographic image dataset for each bone structure in order to rigorously evaluate the efficiency of the proposed method. The proposed method resulted in more accurate segmentations compared to the probabilistic patch-based label fusion model (PB) and the classical patch-based majority voting fusion scheme (MV) using different registration strategies. Comparison with manual (gold standard) segmentations revealed that the good classification accuracy of our unsupervised segmentation scheme is, respectively, 93.79% for the patella, 88.3% for the talus and 85.02% for the pelvis; a score that falls within the range of accuracy levels of manual segmentations (due to the intra inter/observer variability).

Image restoration using functional and anatomical information fusion with application to SPECT-MRI images.

Image restoration is usually viewed as an ill-posed problem in image processing, since there is no unique solution associated with it. The quality of restored image closely depends on the constraints imposed of the characteristics of the solution. In this paper, we propose an original extension of the NAS-RIF restoration technique by using information fusion as prior information with application in SPECT medical imaging. That extension allows the restoration process to be constrained by efficiently incorporating, within the NAS-RIF method, a regularization term which stabilizes the inverse solution. Our restoration method is constrained by anatomical information extracted from a high resolution anatomical procedure such as magnetic resonance imaging (MRI). This structural anatomy-based regularization term uses the result of an unsupervised Markovian segmentation obtained after a preliminary registration step between the MRI and SPECT data volumes from each patient. This method was successfully tested on 30 pairs of brain MRI and SPECT acquisitions from different subjects and on Hoffman and Jaszczak SPECT phantoms. The experiments demonstrated that the method performs better, in terms of signal-to-noise ratio, than a classical supervised restoration approach using a Metz filter.

3D biplanar statistical reconstruction of scoliotic vertebrae.

A new 3D reconstruction method of scoliotic vertebrae of a spine, using two calibrated conventional radiographic images (postero-anterior and lateral), and a global prior knowledge on the geometrical structure of each vertebra is presented. This geometrical knowledge is efficiently captured by a statistical deformable template integrating a set of admissible deformations, expressed by the first modes of variation in the Karhunen-Loeve expansion of the pathological deformations observed on a representative scoliotic vertebra population. The proposed reconstruction method consists in fitting the projections of this deformable template with the preliminary segmented contours of the corresponding vertebra on the two radiographic views. The 3D reconstruction problem is stated as the minimization of a cost function for each vertebra and solved with a gradient descent technique. The reconstruction of the spine is then made vertebra by vertebra. The proposed method allows also to efficiently obtain an accurate 3D reconstruction of each scoliotic vertebra and, consequently, it allows also to get an accurate knowledge of the 3D structure of the whole scoliotic spine. This reconstruction method is in final phase of validation.

A multiscale optimization approach for the dynamic contour-based boundary detection issue.

We present a new multiscale approach for deformable contour optimization. The method relies on a multigrid minimization method and a coarse-to-fine relaxation algorithm. This approach consists in minimizing a cascade of optimization problems of reduced and increasing complexity instead of considering the minimization problem on the full and original configuration space. Contrary to classical multiresolution algorithms, no reduction of image is applied. The family of defined energy functions are derived from the original (full resolution) objective function, ensuring that the same function is handled at each scale and that the energy decreases at each step of the deformable contour minimization process. The efficiency and the speed of this multiscale optimization strategy is demonstrated in the difficult context of the minimization of a region-based contour energy function ensuring the boundary detection of anatomical structures in ultrasound medical imagery. In this context, the proposed multiscale segmentation method is compared to other classical region-based segmentation approaches such as Maximum Likelihood or Markov Random Field-based segmentation techniques. We also extend this multiscale segmentation strategy to active contour models using a classical edge-based likelihood approach. Finally, time and performance analysis of this approach, compared to the (commonly used) dynamic programming-based optimization procedure, is given and allows to attest the accuracy and the speed of the proposed method.

Three-dimensional blind deconvolution of SPECT images.

Thanks to its ability to yield functionally rather than anatomically-based information, the three-dimensional (3-D) SPECT imagery technique has become a great help in the diagnostic of cerebrovascular diseases. Nevertheless, due to the imaging process, the 3-D single photon emission computed tomography (SPECT) images are very blurred and, consequently, their interpretation by the clinician is often difficult and subjective. In order to improve the resolution of these 3-D images and then to facilitate their interpretation, we propose herein to extend a recent image blind deconvolution technique (called the nonnegativity support constraint-recursive inverse filtering deconvolution method) in order to improve both the spatial and the interslice resolution of SPECT volumes. This technique requires a preliminary step in order to find the support of the object to be restored. In this paper, we propose to solve this problem with an unsupervised 3-D Markovian segmentation technique. This method has been successfully tested on numerous real and simulated brain SPECT volumes, yielding very promising restoration results.

Sonar image segmentation using an unsupervised hierarchical MRF model.

This paper is concerned with hierarchical Markov random field (MRP) models and their application to sonar image segmentation. We present an original hierarchical segmentation procedure devoted to images given by a high-resolution sonar. The sonar image is segmented into two kinds of regions: shadow (corresponding to a lack of acoustic reverberation behind each object lying on the sea-bed) and sea-bottom reverberation. The proposed unsupervised scheme takes into account the variety of the laws in the distribution mixture of a sonar image, and it estimates both the parameters of noise distributions and the parameters of the Markovian prior. For the estimation step, we use an iterative technique which combines a maximum likelihood approach (for noise model parameters) with a least-squares method (for MRF-based prior). In order to model more precisely the local and global characteristics of image content at different scales, we introduce a hierarchical model involving a pyramidal label field. It combines coarse-to-fine causal interactions with a spatial neighborhood structure. This new method of segmentation, called the scale causal multigrid (SCM) algorithm, has been successfully applied to real sonar images and seems to be well suited to the segmentation of very noisy images. The experiments reported in this paper demonstrate that the discussed method performs better than other hierarchical schemes for sonar image segmentation.