Carola Fassnacht

Multi-scale line segmentation with automatic estimation of width, contrast and tangential direction in 2D and 3D medical images

A new multi-scale segmentation technique for line-like structures in 2D and 3D medical images is presented. It is based on normalized first and second derivatives and on the eigenvector analysis of the hessian matrix. Application areas are the segmentation and tracking of bloodvessels, electrodes, catheters and other line-like objects. It allows for the estimation of the local diameter, the longitudinal direction and the contrast of the vessel and for the distinction between edge-like and line-like structures. The method is applicable as automatic 2D and 3D line-filter, as well as for interactive algorithms that are based on local direction estimation. A 3D line-tracker has been constructed that uses the estimated longitudinal direction as step-direction. After extraction of the centerline, the hull of the structure is determined by a 2D active-contour algorithm, applied in planes, orthogonal to the longitudinal line-direction. The procedure results in a stack of contours allowing quantitative crosssection area determination and visualization by means of a triangulation based rendering.

An approach to 2D/3D registration of a vertebra in 2D X-ray fluoroscopies with 3D CT images

In order to use pre-operative images during an operation for navigation, they must be registered to the patients coordinate system in the operating theater or to an intra-operative image. One problem in this area is the registration of a vertebra in intra-operatively acquired x-ray fluoroscopies with 3D CT images obtained before the intervention. The result can be used to support the placement of pedicle screws in spine surgery or stents in Transfemoral Endovascular Aneurysm Management (TEAM). For this 2D/3D registration task a novel voxel-based method is presented. Using a small part of the CT image covering the vertebra only, pseudo projections are computed and the resulting vertebra template is compared to the x-ray projection. A new similarity measure was introduced for that purpose, because commonly used measures did not work. The method allows for a much faster implementation than other voxel-based 2D/3D registration approaches, because they use the entire CT image to calculate pseudo projections. Unlike contour-based 2D/3D registration approaches, the method does not require segmentation of the vertebras contours in the x-ray projection. Application and performance of the proposed registration method are demonstrated by application to images of a TEAM procedure.

A Multi-scale Line Filter with Automatic Scale Selection Based on the Hessian Matrix for Medical Image Segmentation

A multi-scale segmentation technique for line-like structures in 2D and 3D medical images is presented. It is based on normalized second derivatives and on the eigenvector analysis of the Hessian matrix. The method allows for the estimation of the local diameter, the longitudinal direction and the contrast of line-structures and for the distinction between edge-like and line-like structures. The characteristics of the method in respect to several analytic line-profiles as well as the influence of neighboring structures and line-bending is discussed. The method is applied to 3D medical images.

Image registration: convex weighting functions for histogram-based similarity measures

Recently the entropy-similarity measure has been introduced for the registration of image pairs prior to subtraction in medical imaging e.g. digital subtraction angiography (DSA). The registration is based on motion-vector fields estimated with a template-matching techniques. The entropy is calculated via weighted grey-value histograms of the difference-image template and measures the degree of histogram dispersion in case of misregistration. In this paper, a generalization of the underlying concept is presented. We prove that any strictly convex function can be used as histogram-weighting function leading to a suitable similarity measure. The quality of the histogram-based measures is compared to other frequently used similarity measures. As a result the energy-similarity measure turns out to be the most suitable measure for template matching. The success of the registration will be demonstrated with a geometrically distorted pair of images taken of the abdomen.

Using an Entropy Similarity Measure to Enhance the Quality of DSA Images with an Algorithm Based on Template Matching

The reduction of motion artifacts arising in DSA requires registration of mask- and contrast image prior to subtraction. An algorithm has been developed consisting of a) partitioning an interactively chosen region-of-interest (ROI) and exclusion of low-contrast partitions, b) assignment of homologous landmarks or control points, c) estimation of parameters of an affine transformation and application of the transformation on the mask image (inside the ROI), and d) subtraction of contrast- and corrected mask image. For assigning homologous landmarks, we use the entropy as similarity measure and compare the results to other frequently used measures.

Medical image segmentation with a 3D nearest neighbor Markov mesh

With the aim of tumor segmentation in magnetic resonance (MR) images, the authors employ a hidden 3D Markov mesh model that has been developed for 3D image segmentation in general and has shown promising results on synthetic image data. The authors model the signal intensity within the non-tumorous area in the form of an equiprobable distribution, and they assume that the tumor is characterized by a Gaussian distribution. The authors introduce a class-specific weight coefficient to the Markov model, with which a clinical user can influence the segmentation result. The novelty of this contribution lies in the combination of a three-dimensional hidden mesh model with interaction possibilities for clinical use of the algorithm.

Image segmentation with a propagator Markov mesh model

We introduce the concept of a propagator function to characterize unilateral Markov fields that fulfil the positivity condition. We show that parameter estimation of a model realization corresponds to the solution of a given set of equations, which can be solved explicitly for a specific propagator function depending exclusively on identities among site labels. For a hidden model, we propose a nonsupervised, iterative scheme comprising labeling and parameter re-estimation steps. Segmentation experiments using a third order Markov mesh, together with a look-ahead algorithm for real-time label estimation, show rapid convergence of the learning algorithm and yield subjectively good results.