Steve De Backer

On the construction of an inter-subject diffusion tensor magnetic resonance atlas of the healthy human brain

Voxel based morphometry (VBM) has been increasingly applied to detect diffusion tensor (DT) image abnormalities in patients for different pathologies. An important requisite for a robust VBM analysis is the availability of a high-dimensional non-rigid coregistration technique that is able to align both the spatial and the orientational DT information. Consequently, there is a need for an inter-subject DTI atlas as a group specific reference frame that also contains this orientational DT information. In this work, a population based DTI atlas has been developed that incorporates such orientational DT information with high accuracy and precision. The proposed methodology for constructing such an atlas is compared with a subject based DTI atlas, in which a single subject is selected as the reference image. Our results demonstrate that the population based atlas framework is more accurate with respect to the underlying diffusion information.

Genetic feature selection combined with composite fuzzy nearest neighbor classifiers for hyperspectral satellite imagery

For high-dimensional data, the appropriate selection of features has a significant effect on the cost and accuracy of an automated classifier. In this paper, a feature selection technique using genetic algorithms is applied. For classification, crisp and fuzzy k-nearest neighbor (kNN) classifiers are compared. Composite fuzzy classifier architectures are investigated. Experiments are conducted on airborne visible/infrared imaging spectrometer (AVIRIS) data, and the results are evaluated in the paper.

Comparing isotropic and anisotropic smoothing for voxel‐based DTI analyses: A simulation study

Voxel‐based analysis (VBA) methods are increasingly being used to compare diffusion tensor image (DTI) properties across different populations of subjects. Although VBA has many advantages, its results are highly dependent on several parameter settings, such as those from the coregistration technique applied to align the data, the smoothing kernel, the statistics, and the post‐hoc analyses. In particular, to increase the signal‐to‐noise ratio and to mitigate the adverse effect of residual image misalignments, DTI data are often smoothed before VBA with an isotropic Gaussian kernel with a full width half maximum up to 16 × 16 × 16 mm3. However, using isotropic smoothing kernels can significantly partial volume or voxel averaging artifacts, adversely affecting the true diffusion properties of the underlying fiber tissue. In this work, we compared VBA results between the isotropic and an anisotropic Gaussian filtering method using a simulated framework. Our results clearly demonstrate an increased sensitivity and specificity of detecting a predefined simulated pathology when the anisotropic smoothing kernel was used. Hum Brain Mapp, 2010.

Fusion and merging of multispectral images with use of multiscale fundamental forms

A new multispectral image wavelet representation is introduced, based on multiscale fundamental forms. This representation describes gradient information of multispectral images in a multiresolution framework. The representation is, in particular, extremely suited for fusion and merging of multispectral images. For fusion as well as for merging, a strategy is described. Experiments are performed on multispectral images, where Landsat Thematic Mapper images are fused and merged with SPOT Panchromatic images. The proposed techniques are compared with wavelet-based techniques described in the literature.

On the construction of a ground truth framework for evaluating voxel-based diffusion tensor MRI analysis methods

Although many studies are starting to use voxel-based analysis (VBA) methods to compare diffusion tensor images between healthy and diseased subjects, it has been demonstrated that VBA results depend heavily on parameter settings and implementation strategies, such as the applied coregistration technique, smoothing kernel width, statistical analysis, etc. In order to investigate the effect of different parameter settings and implementations on the accuracy and precision of the VBA results quantitatively, ground truth knowledge regarding the underlying microstructural alterations is required. To address the lack of such a gold standard, simulated diffusion tensor data sets are developed, which can model an array of anomalies in the diffusion properties of a predefined location. These data sets can be employed to evaluate the numerous parameters that characterize the pipeline of a VBA algorithm and to compare the accuracy, precision, and reproducibility of different post-processing approaches quantitatively. We are convinced that the use of these simulated data sets can improve the understanding of how different diffusion tensor image post-processing techniques affect the outcome of VBA. In turn, this may possibly lead to a more standardized and reliable evaluation of diffusion tensor data sets of large study groups with a wide range of white matter altering pathologies. The simulated DTI data sets will be made available online (http://www.dti.ua.ac.be).

Machine learning study of several classifiers trained with texture analysis features to differentiate benign from malignant soft-tissue tumors in T1-MRI images

To study, from a machine learning perspective, the performance of several machine learning classifiers that use texture analysis features extracted from soft‐tissue tumors in nonenhanced T1‐MRI images to discriminate between malignant and benign tumors.

Denoising of multicomponent images using wavelet least-squares estimators

In this paper, we study denoising of multicomponent images. The presented procedures are spatial wavelet-based denoising techniques, based on Bayesian least-squares optimization procedures, using prior models for the wavelet coefficients that account for the correlations between the spectral bands. We analyze three mixture priors: Gaussian scale mixture models, Bernoulli-Gaussian mixture models and Laplacian mixture models. These three prior models are studied within the same framework of least-squares optimization. The presented procedures are compared to Gaussian prior model and single-band denoising procedures. We analyze the suppression of non-correlated as well as correlated white Gaussian noise on multispectral and hyperspectral remote sensing data and Rician distributed noise on multiple images of within-modality magnetic resonance data. It is shown that a superior denoising performance is obtained when (a) the interband covariances are fully accounted for and (b) prior models are used that better approximate the marginal distributions of the wavelet coefficients.

Wavelet-based feature extraction for hyperspectral vegetation monitoring

The high spectral and high spatial resolution, intrinsic to hyperspectral remote sensing, result in huge quantities of data, which slows down the data processing and can result in a poor performance of classifiers. To improve the classification performance, efficient feature extraction methods are needed. This paper introduces a set of features based on the discrete wavelet transform (DWT). Wavelet coefficients, wavelet energies and wavelet detail histogram features are employed as new features for classification. As a feature reduction procedure, we propose a sequential floating search method. Selection is performed using a cost function based on the estimated probability of error, using the Fisher criterion. This procedure selects the best combination of features. To demonstrate the proposed wavelet features and selection procedure, we apply it to vegetation stress detection. For this application, it is shown that wavelet coefficients outperform spectral reflectance and that the proposed selection procedure outperforms combining the best single features.

Towards fully user transparent task and data parallel image processing

This paper reports on the integration of parallel image processing in the ITK library and on improvements to the state-of-the-art of user transparency. In our approach, image processing tasks are wrapped into objects which are passed to the parallel engine. The engine is able to exploit data and task parallelism when executing the tasks on multicores, clusters and/or GPUs. All features necessary for efficient parallel processing are specified by the task objects. The engine can figure out most of the features itself, and is able to check the correctness of the features provided by the user. Interoperation optimization is attained by efficient scheduling of the tasks. The task dependency graph is automatically created at runtime. This is possible by delaying the execution of the tasks and by the intrinsic ITK pipeline updating mechanism. The low-level functions are also made available for the user, as well as a library-independent version.

Fibered fluorescence microscopy (FFM) of intra epidermal nerve fibers--translational marker for peripheral neuropathies in preclinical research: processing and analysis of the data

Peripheral neuropathy can be caused by diabetes or AIDS or be a side-effect of chemotherapy. Fibered Fluorescence Microscopy (FFM) is a recently developed imaging modality using a fiber optic probe connected to a laser scanning unit. It allows for in-vivo scanning of small animal subjects by moving the probe along the tissue surface. In preclinical research, FFM enables non-invasive, longitudinal in vivo assessment of intra epidermal nerve fibre density in various models for peripheral neuropathies. By moving the probe, FFM allows visualization of larger surfaces, since, during the movement, images are continuously captured, allowing to acquire an area larger then the field of view of the probe. For analysis purposes, we need to obtain a single static image from the multiple overlapping frames. We introduce a mosaicing procedure for this kind of video sequence. Construction of mosaic images with sub-pixel alignment is indispensable and must be integrated into a global consistent image aligning. An additional motivation for the mosaicing is the use of overlapping redundant information to improve the signal to noise ratio of the acquisition, because the individual frames tend to have both high noise levels and intensity inhomogeneities. For longitudinal analysis, mosaics captured at different times must be aligned as well. For alignment, global correlation-based matching is compared with interest point matching. Use of algorithms working on multiple CPUs (parallel processor/cluster/grid) is imperative for use in a screening model.