Refereed conference proceedings

Authors affiliated with CBA are in bold.

1. Cluster Detection and Field-of-View Quality Rating: Applied to Automated Pap-Smear Analysis
   Authors: Marine Astruc (1), Patrik Malm, Rajesh Kumar (2), Ewert Bengtsson
   (1) Ecole Centrale Nantes, France
   (2) Centre for Development of Advanced Computing, Thiruvananthapuram, India
   In Proceedings: 2nd International Conference on Pattern Recognition Applications and Methods (ICPRAM), Barcelona, Spain, pages 355-364
   Abstract: Automated cervical cancer screening systems require high resolution analysis of a large number of epithelial cells, involving complex algorithms, mainly analysing the shape and texture of cell nuclei. This can be a very time consuming process. An initial selection of relevant fields-of-view in low resolution images could limit the number of fields to be further analysed at a high resolution. In particular, the detection of cell clusters is of interest for nuclei segmentation improvement, and for diagnostic purpose, malignant and endometrial cells being more prone to stick together in clusters than other cells. In this paper, we propose methods aiming at evaluating the quality of fields-of-view in bright-field microscope images of cervical cells. The approach consists in the construction of neighbourhood graphs using the nuclei as the set of vertices. Transformations are then applied on such graphs in order to highlight the main structures in the image. The methods result in the delineation of regions with varying cell density and the identification of cell clusters. Clustering methods are evaluated using a dataset of manually delineated clusters and compared to a related work.

2. An Algorithm for Parallel Calculation of Trigonometric and Exponential Functions
   Authors: Tony Barrera (1), Anders Hast, Ewert Bengtsson
   (1) Uppsala, Sweden
   In Proceedings: ACM International Conference on Computing Frontiers, Ischia, Italy, paper 8
   Abstract: We propose a new way of calculating the sine and cosine functions. The method is based on recursive applications of a modified complex power algorithm. On a machine with multiple complex multipliers the method can be used to calculate sines and cosines in logarithmic time. The serial version of the presented method requires only two precomputed constants and no tables. In the parallel versions a trade off can be made between the number of parallel processing elements and the size of tables.

3. A Weight Sequence Distance Function
   Authors: Benedek Nagy (1), Robin Strand, Nicolas Normand (2)
   (1) Department of Computer Science, University of Debrecen, Hungary
   (2) Université de Nantes, France
   In Proceedings: Mathematical Morphology and Its Applications to Signal and Image Processing (ISMM), Uppsala, Sweden, Lecture Notes in Computer Science 7883, pages 292-301
   Abstract: In this paper, a family of weighted neighborhood sequence distance functions defined on the square grid is presented. With this distance function, the allowed weight between any two adjacent pixels along a path is given by a weight sequence. We build on our previous results, where only two or three unique weights are considered, and present a framework that allows any number of weights. We show that the rotational dependency can be very low when as few as three or four unique weights are used. An algorithm for computing the distance transform (DT) that can be used for image processing applications is also presented.

4. Dual B-spline Snake for Interactive Myocardial Segmentation
   Authors: Kevin Bianchi (1), Antoine Vacavant (1), Robin Strand, Pierre Terve (2), Laurent Sarry (1)
   (1) ISIT UMR6284 CNRS, Univ. d'Auvergne, Clermont-Ferrand, France
   (2) KEOSYS Company 1, Saint Herblain, France
   In Proceedings: Medical Image Understanding and Analysis (MIUA), Birmingham, UK
   Abstract: This paper presents a novel interactive segmentation formalism based on two coupled B-Spline snake models to efficiently and simultaneously extract myocardial walls from short-axis magnetic resonance images. The main added value of this model is interaction as it is possible to quickly and intuitively correct the result in complex cases without restarting the whole segmentation working flow. During this process, energies computed from the images guide the user to the best position of the model.

5. Salience-Based Parabolic Structuring Functions
   Authors: Vladimir Curic, Cris L. Luengo Hendriks
   In Proceedings: Mathematical Morphology and Its Applications to Signal and Image Processing, Uppsala, Sweden, Lecture Notes in Computer Science 7883, pages 183-194
   Abstract: It has been shown that the use of the salience map based on the salience distance transform can be useful for the construction of spatially adaptive structuring elements. In this paper, we propose salience-based parabolic structuring functions that are defined for a fixed, predefined spatial support, and have low computational complexity. In addition, we discuss how to properly define adjunct morphological operators using the new spatially adaptive structuring functions. It is also possible to obtain flat adaptive structuring elements by thresholding the salience-based parabolic structuring functions

6. A New Quantitative Approach for Estimating Bone Cell Connections from Nano-CT Images
   Authors: Pei Dong (1), Alexandra Pacureanu, Maria Zuluaga (2), Cecile Olivier (1), Frederique Frouin (3), Quentin Grimal (4), Francoise Peyrin (1)
   (1) European Synchrotron Radiation Facility and CREATIS, Université de Lyon, France
   (2) University College London, UK
   (3) Faculté de Médecine Pierre et Marie Curie - Pitié Salpétrière, Paris, France
   (4) Université Pierre et Marie Curie, Paris, France
   In Proceedings: IEEE 35th Annual International Conference on Engineering in Medicine and Biology Society (EMBC), Osaka, Japan, pages 3694-3697
   Abstract: Recent works highlighted the crucial role of the osteocyte system in bone fragility. The number of canaliculi of osteocyte lacuna (Lc.NCa) is an important parameter that reflects the functionality of bone tissue, but rarely reported due to the limitations of current microscopy techniques, and only assessed from 2D histology sections. Previously, we showed the Synchrotron Radiation nanotomography (SR-nanoCT) is a promising technique to image the 3D lacunar-canalicular network. Here we present, for the first time, an automatic method to quantify the connectivity of bone cells in 3D. After segmentation, our method first separates and labels each lacuna in the network. Then, by creating a bounding surface around lacuna, the Lc.NCa is calculated through estimating 3D topological parameters. The proposed method was successfully applied to a 3D SR-nanoCT image of cortical femoral bone. Statistical results on 165 lacunae are reported, showing a mean of 51, which is consistent with the literature.

7. Epithelial Cell Segmentation in Histological Images of Testicular Tissue Using Graph-Cut
   Authors: Azadeh Fakhrzadeh, Ellinor Spörndly-Nees (1), Lena Holm (1), Cris L. Luengo Hendriks
   (1) Department of Anatomy, Physiology and Biochemistry, SLU, Uppsala
   In Proceedings: 17th International Conference on Image Analysis and Processing, Naples, Italy, Lecture Notes in Computer Science 8157, pages 201-208
   Abstract: Computerized image processing has provided us with valuable tools for analyzing histology images. However, histology images are complex, and the algorithm which is developed for a data set may not work for a new and unseen data set. The preparation procedure of the tissue before imaging can significantly affect the resulting image. Even for the same staining method, factors like delayed fixation may alter the image quality. In this paper we face the challenging problem of designing a method that works on data sets with strongly varying quality. In environmental research, due to the distance between the site where the wild animals are caught and the laboratory, there is always a delay in fixation. Here we suggest a segmentation method based on the structural information of epithelium cell layer in testicular tissue. The cell nuclei are detected using the fast radial symmetry filter. A graph is constructed on top of the epithelial cells. Graph-cut optimization method is used to cut the links between cells of different tubules. The algorithm is tested on five different groups of animals. Group one is fixed immediately, three groups were left at room temperature for 18, 30 and 42 hours respectively, before fixation. Group five was frozen after 6 hours in room temperature and thawed. The suggested algorithm gives promising results for the whole data set.

8. Epithelial Cell Layer Segmentation Using Graph-cut and Its Application in Testicular Tissue
   Authors: Azadeh Fakhrzadeh, Ellinor Spörndly-Nees (1), Lena Holm (1), Cris L. Luengo Hendriks
   (1) Department of Anatomy, Physiology and Biochemistry, SLU, Uppsala
   In Proceedings: Medical Image Understanding and Analysis (MIUA), Birmingham, UK
   Abstract: Computerized image processing has provided us with valuable tools for analyzing histology images. However, histology images are complex, and the algorithm which is developed for a data set may not work for a new and unseen data set. The preparation procedure of the tissue before imaging can significantly affect the resulting image. Even for the same staining method, factors like delayed fixation may alter the image quality. In this paper we face the challenging problem of designing a method that works on data sets with strongly varying quality. In environmental research, due to the distance between the site where the wild animals are caught and the laboratory, there is always a delay in fixation. Here we suggest a segmentation method based on the structural information of epithelium cell layer in testicular tissue. The cell nuclei are detected using the fast radial symmetry filter. A graph is constructed on top of the epithelial cells. Graph-cut optimization method is used to cut the links between cells of different tubules. The algorithm is tested on five different groups of animals. Group one is fixated immediately, four groups were left at room temperature for 6, 18, 30 and 42 hours respectively, before fixation. The suggested algorithm gives promising results for the whole data set.

9. Shortest Diagonal Triangulation of Convex Layers
   Authors: Anders Hast, Peter Jenke (1), Stefan Seipel
   (1) University of Gävle
   In Proceedings: The IASTED International Conference on Signal Processing, Pattern Recognition and Applications, Innsbruck, Austria, pages 1-7
   Abstract: One problem in the field of computational geometry is the triangulation of convex layers. The rotating caliper algorithm is an alternative to the constrained Delaunay triangulation method. We present an improved triangulation algorithm, which gives a mesh quality close to that of the Constrained Delaunay but substantially faster. Each layer will be connected to the neighboring layer by edges and from the two vertices constituting an edge the proposed algorithm will select the shortest diagonal to its next neighbors in the polygonal chain on the other side, i.e. from the outer layer to the inner layer or vice versa. We discuss quality issues regarding the rotating caliper method and some improvements to it, as well as how a Constrained Delaunay can be efficiently implemented for convex layers.

10. Rotation Invariant Feature Matching - Based on Gaussian Filtered Log Polar Transform and Phase Correlation
    Authors: Anders Hast, Andrea Marchetti (1)
    (1) IIT, CNR
    In Proceedings: 8th International Symposium on Image and Signal Processing and Analysis (ISPA), Trieste, Italy, pages 1-6
    Abstract: Rotation invariance is an important property for any feature matching method and it has been implemented in different ways for different methods. The Log Polar Transform has primarily been used for image registration where it is applied after phase correlation, which in its turn is applied on the whole images or in the case of template matching, applied on major parts of them followed by an exhaustive search. We investigate how this transform can be used on local neighborhoods of features and how phase correlation as well as normalized cross correlation can be applied on the result. Thus, the order is reversed and we argue why it is important to do so. We demonstrate a common problem with the log polar transform and that many implementations of it are not suitable for local feature detectors. We propose an implementation of it based on Gaussian filtering. We also show that phase correlation generally will perform better than normalized cross correlation. Both handles illumination differences well, but changes in scale is handled better by the phase correlation approach.

11. Automated Quantification of Zebrafish Tail Deformation for High-Throughput Drug Screening
    Authors: Omer Ishaq, Joseph Negri (1), Mark-Anthony Bray (1), Alexandra Pacureanu, Randall T. Peterson (2) Carolina Wählby (1)
    (1) Broad Institute of Harvard and MIT, Cambridge, MA, USA
    (2) Massachusetts General Hospital, Harvard Medical School, Department of Medicine, Charlestown, MA, USA
    In Proceedings: 10th International Symposium on Biomedical Imaging : From Nano to Macro, San Francisco, CA, USA pages 902-905
    Abstract: Zebrafish (Danio rerio) is an important vertebrate model organism in biomedical research thanks to its ease of handling and translucent body, enabling in vivo imaging. Zebrafish embryos undergo spinal deformation upon exposure to chemical agents that inhibit DNA repair. Automated image-based quantification of spine deformation is therefore attractive for whole-organism based assays for use in early-phase drug discovery. We propose an automated method for accurate high-throughput measurement of tail deformations in multi-fish micro-plate wells. The method generates refined medial representations of partial tail-segments. Subsequently, these disjoint segments are analyzed and fused to generate complete tails. Based on estimated tail curvatures we reach a classification accuracy of 91% on individual animals as compared to known control treatment. This accuracy is increased to 95% when combining scores for fish in the same well.

12. Coverage Segmentation of Thin Structures by Linear Unmixing and Local Centre of Gravity Attraction
    Authors: Kristina Lidayová, Joakim Lindblad, Nataša Sladoje, Hans Frimmel (1)
    (1) Division of Scientific Computing, UU
    In Proceedings: 8th International Symposium on Image and Signal Processing and Analysis, Trieste, Italy, pages 83-88
    Abstract: We present a coverage segmentation method for extracting thin structures in two-dimensional images. These thin structures can be, for example, retinal vessels, or microtubules in cytoskeleton, which are often 1-2 pixels thick. There exist several methods for coverage segmentation, but when it comes to thin and long structures, the segmentation is often unreliable. We propose a method that does not shrink the structures inappropriately and creates a trustworthy segmentation. In addition, as a by-product a high-resolution crisp reconstruction is provided. The method needs a reliable crisp segmentation as an input and uses information from linear unmixing and the crisp segmentation to create a high-resolution crisp reconstruction of the object. After a procedure where holes and protrusions are removed, the high-resolution crisp image is optionally down-sampled back to its original size, creating a coverage segmentation that preserves thin structures.

13. Faster Fuzzy Connectedness via Precomputation
    Authors: Filip Malmberg, Robin Strand
    In Proceedings: Mathematical Morphology and Its Applications to Signal and Image Processing (ISSM), Uppsala, Sweden, Lecture Notes in Computer Science 7883, pages 476-483
    Abstract: We propose a method for accelerating the computation of fuzzy connectedness. The method is based on a precomputation step - the construction of a supervertex graph whose vertices are clusters of image elements. By constructing this supervertex graph in a speci c way, we can perform the bulk of the fuzzy connectedness computations on this graph, rather than on the original image, while guaranteeing exact results. Typically, the number of nodes in the supervertex graph is much smaller than the number of elements in the image, and thus less computation is required. In an experiment, we demonstrate the ability of the proposed method to accelerate the computation of fuzzy connectedness considerably.

14. Digital Distances and Integer Sequences
    Authors: Nicolas Normand (1), Robin Strand
    (1) LUNAM Université, Université de Nantes, France
    In Proceedings: 17th IAPR International Conference on Discrete Geometry for Computer Imagery, Seville, Spain, Lecture Notes in Computer Science 7749, pages 169-179
    Abstract: In recent years, the theory behind distance functions defined by neighbourhood sequences has been developed in the digital geometry community. A neighbourhood sequence is a sequence of integers, where each element defines a neighbourhood. In this paper, we establish the equivalence between the representation of convex digital disks as an intersection of half-planes (-representation) and the expression of the distance as a maximum of non-decreasing functions. Both forms can be deduced one from the other by taking advantage of the Lambek-Moser inverse of integer sequences. Examples with finite sequences, cumulative sequences of periodic sequences and (almost) Beatty sequences are given. In each case, closed-form expressions are given for the distance function and -representation of disks. The results can be used to compute the pair-wise distance between points in constant time and to find optimal parameters for neighbourhood sequences.

15. Precise 3D Angle Measurements in CT Wrist Images
    Authors: Johan Nysjö, Albert Christersson (1), Ida-Maria Sintorn, Ingela Nyström, Sune Larsson (1), Filip Malmberg
    (1) Dept. of Orthopaedics, UU
    In Proceedings: Image Analysis and Processing - ICIAP 2013: Part II, Naples, Italy, Lecture Notes in Computer Science 8157, pages 479-488
    Abstract: The clinically established method to assess the displacement of a distal radius fracture is to manually measure two reference angles,the dorsal angle and the radial angle, in consecutive 2D X-ray images of the wrist. This approach has the disadvantage of being sensitive to operator errors since the measurements are performed on 2D projections of a 3D structure. In this paper, we present a semi-automatic system for measuring relative changes in the dorsal angle in 3D computed tomography (CT) images of fractured wrists. We evaluate the proposed 3D measurement method on 28 post-operative CT images of fractured wrists and compare it with the radiographic 2D measurement method used in clinical practice. The results show that our proposed 3D measurement method has a high intra- and inter-operator precision and is more precise and robust than the conventional 2D measurement method

16. SplineGrip - An Eight Degrees-of-Freedom Flexible Haptic Sculpting Tool
    Authors: Pontus Olsson, Fredrik Nysjö, Björn Aneer (1), Stefan Seipel, Ingrid B. Carlbom
    (1) Independent Artist, Stockholm, Sweden
    In Proceedings: 40th International Conference ACM SIGGRAPH 2013, Anaheim, USA, paper 50
    Abstract: SplineGrip is a flexible haptic sculpting tool that senses the articulation and pose (position and orientation) of the sculpting hand in eight degrees-of-freedom (DOF). The tool captures the hand articulation in two DOF, and uses a commercial haptic device to track the hand pose in six DOF and to simultaneously provide three DOF haptic feedback. The eight DOF input is mapped to the pose and shape of a virtual NURBS-based sculpting tool, offering versatile interaction with a virtual model. We capture the hand articulation in two DOF using two bend sensors with curvature dependent resistance, which are attached in two directions to a flexible plastic sheet mounted on the gimbal of the haptic device. One sensor measures the plastic sheet curvature controlled by the thumb, and the other measures the curvature controlled by the middle and ring fingers. In a neutral state, when all fingers are straight, the virtual sculpting tool takes the shape of a line segment. By bending one sensor with the middle and ring fingers, the user changes the virtual tool curvature. By bending the other sensor with the thumb, the user changes the width of the virtual tool. A curvature increase at zero width turns the line into a spline, and a width increase at zero curvature creates a plane. By bending both sensors, the user may simultaneously control the curvature and width of the NURBS surface. The user may toggle between negative and positive curvatures to make convex and concave tools. We demonstrate SplineGrip with a simple sculpting system where the user starts with a block of material and uses the virtual sculpting tool to gradually remove material; the sculpting tool is not limited to subtractive modeling, but can work with other modeling paradigms.

17. Snap-to-fit, a Haptic 6 DOF Alignment Tool for Virtual Assembly
    Authors: Pontus Olsson, Fredrik Nysjö, Jan-Michaél Hirsch (1), Ingrid B, Carlbom
    (1) Oral and Maxillofacial Surgery, UU
    In Proceedings: IEEE World Haptics (WHC), Daejeon, South Korea, pages 205-210
    Abstract: Virtual assembly of complex objects has application in domains ranging from surgery planning to archaeology. In these domains the objective is to plan the restoration of skeletal anatomy or archaeological artifacts to achieve an optimal reconstruction without causing further damage. While graphical modeling plays a central role in virtual assembly, visual feedback alone is often insufficient since object contact and penetration is difficult to discern due to occlusion. Haptics can improve an assembly task by giving feedback when objects collide, but precise fitting of fractured objects guided by delicate haptic cues similar to those present in the physical world requires haptic display transparency beyond the performance of today's systems. We propose a haptic alignment tool that combines a 6 Degrees of Freedom (DOF) attraction force with traditional 6 DOF contact forces to pull a virtual object towards a local stable fit with a fixed object. The object forces are integrated into a virtual coupling framework yielding a stable haptic tool. We demonstrate the use of our system on applications from both cranio-maxillofacial surgery and archaeology, and show that we can achieve haptic rates for fractured surfaces with over 5000 points.

18. Dual-Domain Visual Exploration of Urban Solar Potential
    Authors: Stefan Seipel, David Lingfors (1), Joakim Widén (1)
    (1) Solid State Physics, Dept. Engineering Sciences, UU
    In Proceedings: Eurographics Workshop on Urban Data Modelling and Visualisation, Girona, Spain, pages 21-24
    Abstract: This project aims to improve the planning and design of solar electricity installations in the urban environment. One major objective of these studies is to enable a highly detailed temporal and spatial analysis of the expected solar yield, which becomes increasingly important for optimal load balance in electric power networks. In our research we develop a 3D simulation model that integrates geographical data and detailed 3D urban models with temporal solar irradiance and climate data. According to our model the predicted solar yield becomes a multi-dimensional function of several design-specific parameters that are interactively explored by a human expert. This project is an interdisciplinary initiative that involves researchers from Energy Systems and from Computer Science at Uppsala University and the University of Gävle. During the first year, a demonstrator system for the interactive exploration of the design parameter space has been developed. Our method and the demonstrator system have been published in two international conferences in 2013. Forthcoming research in this project will concern the refinement and validation of computational models, as well new methods for interactive visual exploration.

19. A Probabilistic Template Model for Finding Macromolecules in MET Volume Images
    Authors: Lennart Svensson, Ida-Maria Sintorn
    In Proceedings: 6th Iberian Conference on Pattern Recognition and Image Analysis (IbPRIA), Madeira, Portugal, Lecture Notes in Computer Science 7887, pages 855-962
    Abstract: We introduce and investigate probabilistic templates with particular focus on the application of protein identification in electron tomography volumes. We suggest to create templates with a weighted averaging operation of several object instances after alignment of an identified subpart. The subpart to be aligned should, ideally, correspond to a rigid and easily identifiable part of the object. The proposed templates enable common rigid template matching methods to also find different shape variations without increasing time complexity in the actual search procedure, since a static template is still used. We present general ideas on how to perform the object instance alignment and look specifically at how to do it for the antibody macromolecule IgG.

20. Feature Weight Optimization and Pruning in Historical Text Recognition
    Authors: Fredrik Wahlberg, Anders Brun
    In Proceedings: 9th International Symposium on Visual Computing, Advances in Visual Computing, Rethymnon, Crete, Greece, Lecture Notes in Computer Science 8034, pages 98-107
    Abstract: In handwritten text recognition, "sliding window'' feature extraction represent the visual information contained in written text as feature vector sequences. In this paper, we explore the parameter space of feature weights in search for optimal weights and feature selection using the coordinate descent method. We report a gain of about 5% AUC performance. We use a public dataset for evaluation and also discuss the effects and limitations of "word pruning," a technique in word spotting that is commonly used to boost performance and save computational time.

21. Feature Space Denoising Improves Word Spotting
    Authors: Fredrik Wahlberg, Anders Brun
    In Proceedings: 2nd International Workshop on Historical Document Imaging and Processing, Washington DC, USA, pages 59-66
    Abstract: Some of the sliding window features commonly used in off-line handwritten text recognition are inherently noisy or sensitive to image noise. In this paper, we investigate the effects of several de-noising filters applied in the feature space and not in the image domain. The purpose is to target the intrinsic noise of these features, stemming from the complex shapes of handwritten characters. This noise is present even if the image has been captured without any kind of artefacts or noise. An evaluation, using a public database, is presented showing that the recognition of word-spotting can be improved considerably by using de-noising filters in the feature space.