Biomedical subjects

Frederik Maes

Publications and source records attributed to Frederik Maes.

26 records · Page 2Linked to original sources

PubMed2003

Non-rigid image registration using a statistical spline deformation model.

We propose a statistical spline deformation model (SSDM) as a method to solve non-rigid image registration. Within this model, the deformation is expressed using a statistically trained B-spline deformation mesh. The model is trained by principal component analysis of a training set. This approach allows to reduce the number of degrees of freedom needed for non-rigid registration by only retaining the most significant modes of variation observed in the training set. User-defined transformation components, like affine modes, are merged with the principal components into a unified framework. Optimization proceeds along the transformation components rather then along the individual spline coefficients. The concept of SSDM's is applied to the temporal registration of thorax CR-images using pattern intensity as the registration measure. Our results show that, using 30 training pairs, a reduction of 33% is possible in the number of degrees of freedom without deterioration of the result. The same accuracy as without SSDM's is still achieved after a reduction up to 66% of the degrees of freedom.

Algorithms↗

PubMed2003

Accuracy of diffusion-weighted MR imaging in the diagnosis of sporadic Creutzfeldt-Jakob disease.

The definitive diagnosis of sporadic Creutzfeldt-Jakob disease (sCJD) is based on brain autopsy. The 14-3-3 analysis in the CSF is considered a highly sensitive and specific procedure. Sensitivity, specificity and accuracy of EEG, the 14-3-3 assay and MR imaging in 12 patients referred for suspected sCJD were calculated. We suggest that diffusion-weighted MR imaging (DWI) should be included in the array of diagnostic tests because of the 100 % sensitivity and specificity.

Aged↗

PubMed2003

A viscous fluid model for multimodal non-rigid image registration using mutual information.

We propose a multimodal free-form registration algorithm based on maximization of mutual information. The warped image is modeled as a viscous fluid that deforms under the influence of forces derived from the gradient of the mutual information registration criterion. Parzen windowing is used to estimate the joint intensity probability of the images to be matched. The method is evaluated for non-rigid inter-subject registration of MR brain images. The accuracy of the method is verified using simulated multi-modal MR images with known ground truth deformation. The results show that the root mean square difference between the recovered and the ground truth deformation is smaller than 1 voxel. We illustrate the application of the method for atlas-based brain tissue segmentation in MR images in case of gross morphological differences between atlas and patient images.

Algorithms↗

PubMed2003

A unifying framework for partial volume segmentation of brain MR images.

Accurate brain tissue segmentation by intensity-based voxel classification of magnetic resonance (MR) images is complicated by partial volume (PV) voxels that contain a mixture of two or more tissue types. In this paper, we present a statistical framework for PV segmentation that encompasses and extends existing techniques. We start from a commonly used parametric statistical image model in which each voxel belongs to one single tissue type, and introduce an additional downsampling step that causes partial voluming along the borders between tissues. An expectation-maximization approach is used to simultaneously estimate the parameters of the resulting model and perform a PV classification. We present results on well-chosen simulated images and on real MR images of the brain, and demonstrate that the use of appropriate spatial prior knowledge not only improves the classifications, but is often indispensable for robust parameter estimation as well. We conclude that general robust PV segmentation of MR brain images requires statistical models that describe the spatial distribution of brain tissues more accurately than currently available models.

Algorithms↗

PubMed2002

Dry preparation for virtual CT colonography with fecal tagging using water-soluble contrast medium: initial results.

The purpose of this study was to evaluate the feasibility of a dry bowel preparation, i.e. without laxative fluids, for virtual CT colonography (VCTC), and its impact on patient acceptance compared with conventional colonoscopy (CC). A randomly chosen patient population scheduled for CC ( n=11) was first submitted to VCTC after a dry preparation, consisting of low-residue meals combined with a small amount of a iodinated water-soluble contrast medium during each meal 3 days before VCTC. In different colon segments and between different persons, the degree of tagging in VCTC was evaluated and graded. Patient acceptance and future preference were assessed for both preparations as well as for both investigations. The mixing of the contrast with the intestinal content results in contrast impregnated stool, the tagged feces. The degree of fecal tagging was good in the majority of the patients and the colonic segments, especially in the descending colon and sigmoid. Furthermore, patient acceptance and preference were clearly in favour of VCTC compared with CC merely because of the non-invasiveness of the dry preparation. Dry bowel preparation and VCTC is a promising approach towards a patient-friendly colon cancer-screening setup.

Adult↗

PubMed2002

Evaluation of image features and search strategies for segmentation of bone structures in radiographs using Active Shape Models.

In this paper, we evaluate various image features and different search strategies for fitting Active Shape Models (ASM) to bone object boundaries in digitized radiographs. The original ASM method iteratively refines the pose and shape parameters of the point distribution model driving the ASM by a least squares fit of the shape to update the target points at the estimated object boundary position, as determined by a suitable object boundary criterion. We propose an improved search procedure that is more robust against outlier configurations in the boundary target points by requiring subsequent shape changes to be smooth, which is imposed by a smoothness constraint on the displacement of neighbouring target points at each iteration and implemented by a minimal cost path approach. We compare the original ASM search method and our improved search algorithm with a third method that does not rely on iteratively refined target point positions, but instead optimizes a global Bayesian objective function derived from statistical a priori contour shape and image models. Extensive validation of these methods on a database containing more than 400 images of the femur, humerus and calcaneus using the manual expert segmentation as ground truth shows that our minimal cost path method is the most robust. We also evaluate various measures for capturing local image appearance around each boundary point and conclude that the Mahalanobis distance applied to normalized image intensity profiles extracted normal to the shape is the most suitable criterion among the tested ones for guiding the ASM optimization.

Algorithms↗

PubMed2002

Retrospective correction of the heel effect in hand radiographs.

A method for the retrospective correction of intensity inhomogeneities induced by the heel effect in digital radiographs is presented. The method is based on a mathematical model for the heel effect derived from the acquisition geometry. The model parameters are estimated by fitting the model to the image intensity data in the background or direct exposure area only where the heel effect is directly measurable, while the correction is then applied to the whole image. The method iterates between background segmentation and heel effect correction until convergence. We illustrate the performance of the method on flat field and phantom images and demonstrate its robustness on a database of 137 diagnostic hand radiographs.

Algorithms↗

PubMed2002

Improved visualization of coronary arteries using a new three-dimensional submillimeter MR coronary angiography sequence with balanced gradients.

OBJECTIVE: The goal of our study was to evaluate a new three-dimensional real-time navigator MR coronary angiography sequence to noninvasively visualize the coronary arteries. SUBJECTS AND METHODS: Fifteen healthy volunteers underwent MR coronary angiography with a new balanced turbo field-echo sequence in comparison with the standard turbo field-echo sequence. Signal-to-noise, blood-to-myocardium, blood-to-fat, and blood-to-pericardial fluid contrast ratios of the left and right coronary artery systems were measured. Image quality was graded, the length and diameter of the coronary arteries were measured, and the number of visible side branches was assessed. RESULTS: The balanced turbo field-echo images yielded a higher blood-to-myocardium and blood-to-pericardial fluid contrast ratio, a similar blood-to-fat contrast ratio, and a lower signal-to-noise ratio than the turbo field-echo images. On a 5-point grading scale (1, nondiagnostic or unreadable; 2, poor; 3, moderate; 4, good; 5, excellent), image quality was rated significantly better for the balanced turbo field-echo sequence than for the turbo field-echo sequence (left coronary artery, 4.0 +/- 0.6 vs 3.6 +/- 0.5 [p = 0.015]; right coronary artery, 4.4 +/- 0.4 vs 3.6 +/- 0.4 [p < 0.0001], respectively), resulting in a significantly longer segment of the three major coronary arteries visualized (left anterior descending coronary artery, 92 +/- 21 mm vs 79 +/- 24 mm; left circumflex coronary artery, 70 +/- 7 mm vs 60 +/- 18 mm; right coronary artery, 112 +/- 28 mm vs 95 +/- 27 mm) and a significantly higher number of side branches visualized (left anterior descending coronary artery, 2.9 +/- 1.3 vs 1.5 +/- 1.3; left circumflex coronary artery, 2.1 +/- 1.7 vs 1.0 +/- 1.2; right coronary artery, 3.7 +/- 1.7 vs 2.6 +/- 1.5). Mean imaging time per coronary artery was significantly shorter for the balanced turbo field-echo sequence (5.7 +/- 1.0 min) than for the turbo field-echo sequence (8.4 +/- 1.4 min) (p < 0.0001). CONCLUSION: Compared with standard turbo field-echo MR coronary angiography, optimized balanced turbo field-echo MR coronary angiography improves the visualization of the coronary arteries and their side branches within a significantly shorter imaging time.

Adult↗