Refereed conference proceedings

1. Recognizing signs of malignancy: The quest for computer assisted cancer screening and diagnosis systems
   Authors: Ewert Bengtsson
   Conference: IEEE International Conference on Computational Intelligence and Computing Research (ICCIC), 2010, Coimbatore, India, December 28-29, pp. 1-6
   Publisher: IEEE Computer Society
   Abstract: Almost all cancers are diagnosed through visual examination of microscopic tissue samples. Visual screening of cell samples, so called PAP-smears, has drastically reduced the incidence of cervical cancers in countries that have implemented population wide screening programs. But the visual examination is tedious, subjective and expensive. There has therefore been much research aiming for computer assisted or automated cell image analysis systems for cancer detection and diagnosis. Progress has been made but still most of cytology and pathology is done visually. In this presentation I will discuss some of the major issues involved, examine some of the proposed solutions and give some comments about the state of the art.

2. Three-Dimensional Tracing of Neurites in Fluorescence Microscopy Images Using Local Path-Finding
   Authors: Magnus Gedda and Pascal Vallotton(1)
   (1) CSIRO Mathematical and Information Sciences, Sydney, Australia
   Conference: IEEE International Conference On Acoustics, Speech And Signal Processing (ICASSP), Dallas, Texas, USA, March 14-19, pp. 646-649
   Publisher: IEEE Computer Society
   Abstract: Neurite tracing in 3D neuron images is important when it comes to analysing the growth and functionality of nerve cells. The methods used today are either of high complexity, limiting throughput, or semi-automatic, i.e., requiring user interaction. This makes them unsuitable for analysis where high throughput is needed. In this work we propose a method designed for low complexity and void of user interaction by using local path-finding. The method is illustrated on both phantom and real data, and compared with a widely used commercial software package with promising results.

3. A Cost-Efficient and Automatic Digitization Workflow Using Commodity Hardware and Image Analysis
   Authors: Henrik Johansson(1), Per Erik Svedlund(1), Erik Siira(1), Hamid Sarve
   (1) National Library of Sweden
   Conference: Archiving, Den Haag, The Netherlands, June 1-4, pp. 101-106
   Publisher: Society for Imaging Science and Technology
   Abstract: Digitization projects put large strains on organizations with limited financial resources. Dedicated digitization equipment is expensive, both to acquire and to maintain. Furthermore, a significant workforce is needed to operate the equipment and to perform necessary post-processing tasks. To achieve high digitization throughput with limited resources, both the cost of the digitization equipment and the amount of manual labour must be reduced.

   In this paper, we present the digitization equipment and the digitization workflow at the National Library of Sweden (Kungliga Biblioteket, the KB). The workflow is designed towards a single goal; to achieve the highest possible digitization throughput using the least amount of resources - without any significant loss of quality or risk of compromising delicate objects.

   For the image capture, we use a commodity DSLRs, the Canon 5D Mark II. To make the image capture more efficient, the labour intensive and error-prone tasks of organizing and renaming the image files are automatically performed by in-house software.

   During post-processing, we derive files suitable for presentation. For the presentation files, we remove excess areas around the object and we color correct, resize and sharpen the images. These tasks are time-consuming to perform manually. Hence, we have developed an in-house application to automatically perform the post-processing with minimal input from the operator.

   With the presented workflow, involving commodity hardware and extensive use of in-house software, we have been able to more than triple our digitization throughput despite limited financial resources. The pending use of edge-detection, image matching, and distributed computing will further increase the throughput without the need of additional resources.

4. De-noising of SRµCT Fiber Images by Total Variation Minimization
   Authors: Joakim Lindblad, Nataša Sladoje(1), Tibor Lukic(1)
   (1) University of Novi Sad, Faculty of Engineering
   Conference: 20th International Conference on Pattern Recognition (ICPR), Istanbul, Turkey, August 23-26, pp. 4621-4624
   Publisher: IEEE Computer Society
   Abstract: SRµCT images of paper and pulp fiber materials are characterized by a low signal to noise ratio. De-noising is therefore a common preprocessing step before segmentation into fiber and background components. We suggest a de-noising method based on total variation minimization using a modified Spectral Conjugate Gradient algorithm. Quantitative evaluation performed on synthetic 3D data and qualitative evaluation on real 3D paper fiber data confirm appropriateness of the suggested method for the particular application.

5. PAPSYNTH: Simulated Bright-Field Images of Cervical Smears
   Authors: Patrik Malm, Anders Brun, Ewert Bengtsson
   Conference: IEEE International Symposium on Biomedical Imaging: From Nano to Macro (ISBI), Rotterdam, The Netherlands, April 14-17, pp. 117-120
   Publisher: IEEE Computer Society
   Abstract: In this paper, we present a simulator for bright-field microscope images of ``Pap-smears'', which is the most common technique used today for cervical cancer screening. Lacking a ground truth for real images, these realistic synthetic images may be used to tune and validate image analysis and processing algorithms. We demonstrate this for two tasks: uncorrelated noise removal and nucleus segmentation. The simulator is a part of a larger project, aiming at automatic, cost efficient screening for cervical cancer in developing countries.

6. Two Non-linear Parametric Models of Contrast Enhancement for DCE-MRI of the Breast Amenable to Fitting Using Linear Least Squares
   Author: Andrew Mehnert(1), Michael Wildermoth(1), Stuart Crozier(1), Ewert Bengtsson, Dominic Kennedy(2)
   (1) School of ITEE, University of Queensland, Australia
   (2) Queensland X-Ray, Greenslopes Private Hospital, Australia
   Conference: International Conference on Digital Image Computing: Techniques and Applications (DICTA), Sydney, Australia, December 1-3, pp. 611-616
   Publisher: IEEE Computer Society
   Abstract: This paper proffers two non-linear empirical parametric models-linear slope and Ricker-for use in characterising contrast enhancement in dynamic contrast enhanced (DCE) MRI. The advantage of these models over existing empirical parametric and pharmacokinetic models is that they can be fitted using linear least squares (LS). This means that fitting is quick, there is no need to specify initial parameter estimates, and there are no convergence issues. Furthermore the LS fit can itself be used to provide initial parameter estimates for a subsequent NLS fit (self-starting models). The results of an empirical evaluation of the goodness of fit (GoF) of these two models, measured in terms of both MSE and , relative to a two-compartment pharmacokinetic model and the Hayton model are also presented. The GoF was evaluated using both routine clinical breast MRI data and a single high temporal resolution breast MRI data set. The results demonstrate that the linear slope model fits the routine clinical data better than any of the other models and that the two parameter self-starting Ricker model fits the data nearly as well as the three parameter Hayton model. This is also demonstrated by the results for the high temporal data and for several temporally sub-sampled versions of this data.

7. A Modified Particle Swarm Optimization Applied in Image Registration
   Author: Muhammad Khalid Khan Niazi, Ingela Nyström
   Conference: 20th International Conference on Pattern Recognition (ICPR), Istanbul, Turkey, August 23-26, pp. 2303-2305
   Publisher: IEEE Computer Society
   Abstract: We report a modified version of the particle swarm optimization (PSO) algorithm and its application to image registration. The modified version utilizes benefits from the Gaussian and the uniform distribution, when updating the velocity equation in the PSO algorithm. Which one of the distributions is selected depends on the direction of the cognitive and social components in the velocity equation. This direction checking and selection of the appropriate distribution provide the particles with an ability to jump out of local minima. The registration results achieved by this new version proves the robustness and its ability to find a global minimum.

8. Coupling Visualization and Data Analysis for Knowledge Discovery from Multi-dimensional Scientific Data
   Authors: Oliver Rbel(1), Sean Ahern(2), Bethel, E. Wes(1), Biggin, Mark D.(1), Childs, Hank(1), Estelle, Cormier-Michel(3), Angela, DePace(4), Michael B. Eisen(5), Charless C. Fowlkes(6), Cameron G.R. Geddes(1), Hans Hagen(7), Bernd Hamann(1), Min-Yu Huang(8), Soile V.E. Keränen(1), David W. Knowles(1), Cris L. Luengo Hendriks, Jitendra Malik(5), Jeremy Meredith(2), Peter Messmer(3), Prabhat(1), Daniela Ushizima(1), Gunther H. Weber(1) and Kesheng Wu(1)
   (1) Lawrence Berkeley National Laboratory, California
   (2) Oak Ridge National Laboratory, Tennessee
   (3) Tech-X Corporation, Colorado
   (4) Harvard Medical School, Massachusetts
   (5) University of California, Berkeley
   (6) University of California, Irvine
   (7) University of Kaiserslautern, Germany
   (8) University of California, Davis
   Conference: International Conference on Computational Science (ICCS), Amsterdam, The Netherlands, May 31-June 2
   Procedia Computer Science, 1(1), pp. 1751-1758
   Publisher: Elsevier
   Abstract: Knowledge discovery from large and complex scientific data is a challenging task. With the ability to measure and simulate more processes at increasingly finer spatial and temporal scales, the growing number of data dimensions and data objects presents tremendous challenges for effective data analysis and data exploration methods and tools. The combination and close integration of methods from scientific visualization, information visualization, automated data analysis, and other enabling technologies - such as efficient data management - supports knowledge discovery from multi-dimensional scientific data. This paper surveys two distinct applications in developmental biology and accelerator physics, illustrating the effectiveness of the described approach.

9. Methods for Visualization of Bone Tissue in the Proximity of Implants
   Authors: Hamid Sarve, Joakim Lindblad, Carina Johansson(1), Gunilla Borgefors
   (1) Örebro University
   Conference: International Conference on Computer Vision and Graphics (ICCVG), Warsaw, Poland, September 20-22
   Lecture Notes in Computer Science (LNCS) 6375, pp. 243-250
   Editors: L. Bolc, R. Tadeusiewicz, L.J. Chmielewski, K. Wojciechowski
   Publisher: Springer-Verlag Berlin, Heidelberg
   Abstract: In this work we present two methods for visualization ofSRµ CT-scanned 3D volumes of screw-shaped bone implant samples:thread fly-through and 2D unfolding. The thread fly-through generates an animation by following the thread helix and extracting slices along it. Relevant features, such as bone ratio and bone implant contact, are computed for each slice of the animation and displayed as graphs beside the animation. The 2D unfolding, on the other hand, maps the implantsurface onto which feature information is projected to a 2D image, providing an instant overview of the whole implant. The unfolding is made area-preserving when appropriate. These visualization methods facilitate better understanding of the bone-implant integration and provides a good platform for communication between involved experts.

10. A Local Curvature Based Lighting Model for Rendering of Snow
    Authors: Stefan Seipel and Anders Hast
    Conference: IADIS Computer Graphics, Visualization, Computer Vision and Image Processing Conference (CGVCVIP), Freiburg, Germany, July 27-29, pp. 367-372
    Publisher: International Association for Development of the Information Society
    Comment: Short paper
    Abstract: Local illumination models try to describe the interaction between light and objects in the scene based on only few parameters representing some geometric and material properties at a given point on a surface. In this paper we present our research in local illumination models for the purpose of approximating light transport and shadow-masking effects in the local neighborhood of the surface point under evaluation. We assume that the amount of curvature at a surface point to some extent represents geometric properties in the surrounding neighborhood of this point. From this we define an empirical illumination model for diffuse reflecting materials which controls the amount of locally scattered light from the neighborhood as well as subsurface light transport based on some curvature metric.

11. Sampling and Ideal Reconstruction on the 3D Diamond Grid
    Author: Robin Strand
    Conference: 20th International Conference on Pattern Recognition (ICPR), Istanbul, Turkey, August 23-26, pp. 4609-4612
    Publisher: IEEE Computer Society
    Abstract: This paper presents basic, yet important, properties that can be used when developing methods for image acquisition, processing, and visualization on the diamond grid. The sampling density needed to reconstruct a band-limited signal and the ideal interpolation function on the diamond grid are derived.

12. Interpolation and Sampling on a Honeycomb Lattice
    Author: Robin Strand
    Conference: 20th International Conference on Pattern Recognition (ICPR), Istanbul, Turkey, August 23-26, pp. 2222-2225
    Publisher: IEEE Computer Society
    Abstract: In this paper, we focus on the three-dimensional honeycomb point-lattice in which the Voronoi regions are hexagonal prisms. The ideal interpolation function is derived by using a Fourier transform of the sampling lattice. From these results, the sampling efficiency of the lattice follows.

13. Estimation of Linear Deformations of 3D Objects
    Authors: Attila Tanacs(1), Joakim Lindblad, Nataša Sladoje(2), Zoltan Kato(1)
    (1) Dept. of Image Processing and Computer Graphics, University of Szeged, Szeged, Hungary
    (2) Faculty of Technical Sciences, University of Novi Sad, Serbia
    Conference: 17th International Conference on Image Processing (ICIP), Hong Kong, China, September 26-29, pp. 153-156
    Publisher: IEEE Computer Society
    Abstract: We propose a registration method to find affine transformations between 3D objects by constructing and solving an overdetermined system of polynomial equations. We utilize voxel coverage information for more precise object boundary description. An iterative solution enables us to easily adjust the method to recover e.g. rigid-body and similarity transformations. Synthetic tests show the advantage of the voxel coverage representation, and reveal the robustness properties of our method against different types of segmentation errors. The method is tested on a real medical CT volume.

14. On the Quality of Point Set Triangulations based on Convex Hulls
    Authors: Peter Jenke(1), Anders Hast, Stefan Seipel
    (1) University of Gävle
    Conference: SIGRAD, Swedish Chapter of Eurographics, Västerås, November 25-26, pp. 71-74
    Editors: Kai-Mikael Jää-Aro and Thomas Larsson
    Publisher: Linkping University Electronic Press, Linkping University
    Abstract: In this paper we describe a method for directly generating triangle strips from unstructured point clouds based on onion peeling triangulation (OPT). It is an iterative reconstruction of the convex hulls of point clouds in the 2D plane, and it uses pairs of subsequent layers to establish triangle strips. We compare the obtained triangulations with the results of Delaunay triangulations in terms of the distribution of the symmetry of obtained triangles and in regard to the number of polygons/vertices emitted. Our initial results show that onion peeling is a straightforward method to directly obtain large triangle strips of point clouds. As expected, the triangulation is not as well behaved as in Delaunay-triangulation [VK07]. In terms of triangle complexity and average strip length OPT is a very favorable triangulation alternative which also lends suitable for the triangulation of 3D point clouds.