Skip to main content

About this book

The book discusses novel visualization techniques driven by the needs in medicine and life sciences as well as new application areas and challenges for visualization within these fields. It presents ideas and concepts for visual analysis of data from scientific studies of living organs or to the delivery of healthcare. Target scientific domains include the entire field of biology at all scales - from genes and proteins to organs and populations - as well as interdisciplinary research based on technological advances such as bioinformatics, biomedicine, biochemistry, or biophysics. Moreover, they comprise the field of medicine and the application of science and technology to healthcare problems. This book does not only present basic research pushing the state of the art in the field of visualization, but it also documents the impact in the fields of medicine and life sciences.

Table of Contents


Segmentation and Uncertainty


Lung Segmentation of MR Images: A Review

Magnetic resonance imaging (MRI) is a non-radiation based examination method, which gains an increasing popularity in research and clinical settings. Manual analysis of large data volumes is a very time-consuming and tedious process. Therefore, automatic analysis methods are required. This paper reviews different methods that have been recently proposed for automatic and semi-automatic lung segmentation from magnetic resonance imaging data. These techniques include thresholding, region growing, morphological operations, active contours, level sets, and neural networks. We also discuss the methodologies that have been utilized for performance and accuracy evaluation of each method.
Tatyana Ivanovska, Katrin Hegenscheid, René Laqua, Sven Gläser, Ralf Ewert, Henry Völzke

Fast Uncertainty-Guided Fuzzy C-Means Segmentation of Medical Images

Image segmentation is a crucial step of the medical visualization pipeline. In this paper, we present a novel fast algorithm for modified fuzzy c-means segmentation of MRI data. The algorithm consists of two steps, which are executed as two iterations of a fuzzy c-means approach: the first iteration is a standard fuzzy c-means (FCM) iteration, while the second iteration is our modified FCM iteration with misclassification correction. In the second iteration, we use the classification probability vectors (uncertainties) of the neighbor pixels found by the first iteration to confirm or correct the classification decision of the current pixel. The application of the proposed algorithm on synthetic data, simulated MRI data, and real MRI data show that our algorithm is insensitive to different types of noise and outperforms the standard FCM and several versions of modified FCM algorithms in terms of accuracy and speed. In fact, our algorithm can easily be combined with many modified FCM algorithms to improve their segmentation result while reducing the computation costs (using two FCM iterations only). An optional simple post-processing step can further improve the segmentation result by correcting isolated misclassified pixels. We also show that our algorithm reduces the uncertainty in the segmentation result, by using recently proposed uncertainty estimation and visualization tools.
Ahmed Al-Taie, Horst K. Hahn, Lars Linsen

muView: A Visual Analysis System for Exploring Uncertainty in Myocardial Ischemia Simulations

In this paper we describe the Myocardial Uncertainty Viewer (muView or μView) system for exploring data stemming from the simulation of cardiac ischemia. The simulation uses a collection of conductivity values to understand how ischemic regions effect the undamaged anisotropic heart tissue. The data resulting from the simulation is multi-valued and volumetric, and thus, for every data point, we have a collection of samples describing cardiac electrical properties. μView combines a suite of visual analysis methods to explore the area surrounding the ischemic zone and identify how perturbations of variables change the propagation of their effects. In addition to presenting a collection of visualization techniques, which individually highlight different aspects of the data, the coordinated view system forms a cohesive environment for exploring the simulations. We also discuss the findings of our study, which are helping to steer further development of the simulation and strengthening our collaboration with the biomedical engineers attempting to understand the phenomenon.
Paul Rosen, Brett Burton, Kristin Potter, Chris R. Johnson

Visualization of 3D Medical Images


Combined Volume Registration and Visualization

We describe a method for combining and visualizing a set of overlapping volumetric data sets with high resolution but limited spatial extent. Our system combines the calculation of a registration metric with ray casting for direct volume rendering on the graphics processing unit (GPU). We use the simulated annealing algorithm to find a registration close to optimal and allow the user to closely monitor the optimization progress. The combined calculation reduces memory traffic, increases rendering frame rate, and makes possible interactive-speed, user-supervised, semi-automatic combination of many component volumetric data sets.
Arlie G. Capps, Robert J. Zawadzki, John S. Werner, Bernd Hamann

Feature Lines for Illustrating Medical Surface Models: Mathematical Background and Survey

This paper provides a tutorial and survey for a specific kind of illustrative visualization technique: feature lines. We examine different feature line methods. For this, we provide the differential geometry behind these concepts and adapt this mathematical field to the discrete differential geometry. All discrete differential geometry terms are explained for triangulated surface meshes. These utilities serve as basis for the feature line methods. We provide the reader with all knowledge to re-implement every feature line method. Furthermore, we summarize the methods and suggest a guideline for which kind of surface which feature line algorithm is best suited. Our work is motivated by, but not restricted to, medical and biological surface models.
Kai Lawonn, Bernhard Preim

Remote Visualization Techniques for Medical Imaging Research and Image-Guided Procedures

There has been a tremendous increase in medical image computing research and development over the last decade. This trend continues to gain further speed, driven by the sheer amount of multimodal medical image data but also by the broad spectrum of computer-assisted applications. At the same time, user expectations with respect to diagnostic accuracy, robustness, speed, automation, workflow efficiency, broad availability, as well as intuitive use have reached a high level already. More recently, cloud computing has entered the field of medical imaging, providing means for more flexible workflows including the support of mobile devices and even a medical imaging equivalent of the App Store paradigm. This paper discusses requirements for modern medical software systems with a focus on image analysis and visualization. It provides examples from different areas of application covering collaborative multi-center imaging trials with online reading and advanced analysis as well as an intraoperative augmented-reality scenario for translating liver surgery planning data directly into the operating room through a mobile multi-touch device. A combination of remote rendering and visualization techniques with an efficient modular development framework (MeVisLab) is presented as a basis for fast implementation, early evaluation, and iterative optimization in these applications.
Peter Kohlmann, Tobias Boskamp, Alexander Köhn, Christian Rieder, Andrea Schenk, Florian Link, Uwe Siems, Marcus Barann, Jan-Martin Kuhnigk, Daniel Demedts, Horst K. Hahn

Visualization for Diffusion-Weighted Imaging


Visualization of MRI Diffusion Data by a Multi-Kernel LIC Approach with Anisotropic Glyph Samples

In diffusion weighted magnetic resonance imaging (DW-MRI), high angular resolution imaging techniques have become available, allowing a voxel’s diffusion profile to be measured and represented with high fidelity by a fiber orientation distribution function (FOD), even in situations of crossing and branching white matter fibers. Fiber tractography algorithms, such as streamline tracking, are used for visualizing global relationships between brain regions. However, they are prone to errors, e.g., may miss to visualize relevant fiber branches or provide incorrect connections. Line integral convolution (LIC), when applied to diffusion datasets, yield a more local representation of white matter patterns, and due to the local restriction of its convolution kernel is less susceptible to visualizing erroneous structures. In this paper we propose a multi-kernel LIC approach, which uses anisotropic glyph samples as an input pattern. Derived from FOD functions, multi-cylindrical glyph samples are generated by analysis of a highly-resolved FOD field. This provides a new sampling scheme for the anisotropic packing of samples along integrated fiber lines. Based on this input pattern two- and three-dimensional LIC maps can be constructed, depicting fiber structures with excellent contrast and resolving crossing and branching fiber pathways. We evaluate our approach by simulated DW-MRI data as well as in vivo studies with a healthy volunteer and a brain tumor patient.
Mark Höller, Uwe Klose, Samuel Gröschel, Kay-M. Otto, Hans-H. Ehricke

Exploring Crossing Fibers of the Brain’s White Matter Using Directional Regions of Interest

Diffusion magnetic resonance imaging (dMRI) is a medical imaging method that can be used to acquire local information about the structure of white matter pathways within the human brain. By applying computational methods termed fiber tractography on dMRI data, it is possible to estimate the location and extent of respective nerve bundles (white matter pathways). Visualizing these complex white matter pathways for neuro applications is still an open issue. Hence, interactive visualization techniques to explore and better understand tractography data are required. In this paper, we propose a new interaction technique to support exploration and interpretation of white matter pathways. Our application empowers the user to interactively manipulate manually segmented, box- or ellipsoid-shaped regions of interest (ROIs) to selectively display pathways that pass through specific anatomical areas. To further support flexible ROI design, each ROI can be assigned a Boolean logic operator and a fiber direction. The latter is particularly relevant for kissing, crossing or fanning regions, as it allows the neuroscientists to filter fibers according to their direction within the ROI. By precomputing all white matter pathways in the whole brain, interactive ROI placement and adjustment are possible. The proposed fiber selection tool provides ultimate flexibility and is an excellent approach for fiber tract selection, as shown for some real-world examples.
Andreas Graumann, Mirco Richter, Christopher Nimsky, Dorit Merhof

Multi-Modal Visualization of Probabilistic Tractography

Neuroscientists use visualizations of diffusion data to analyze neural tracts of the brain. More specifically, probabilistic tractography algorithms are a group of methods that reconstruct tract information in diffusion data and need proper visualization. One problem neuroscientists are facing with probabilistic data is putting this information into context. Neuroscience experts already successfully utilized several techniques together with structural MRI to detect neural tracts in the living human brain which were previously only known from tracer studies in macaque monkeys. Whereas the combination with structural MRI, i.e., T1 and T2 images, has been important for these studies, new challenges ask for an integration of other imaging modalities. First, we provide an overview of the currently used visualization techniques. Then, we show how probabilistic tractography can be combined with other techniques, trying to find new and useful visualizations for multi-modal data.
Mathias Goldau, Mario Hlawitschka

Cohort Studies and Time-Varying Phenomena


Visual Analytics of Image-Centric Cohort Studies in Epidemiology

Epidemiology characterizes the influence of causes to disease and health conditions of defined populations. Cohort studies are population-based studies involving usually large numbers of randomly selected individuals and comprising numerous attributes, ranging from self-reported interview data to results from various medical examinations, e.g., blood and urine samples. Since recently, medical imaging has been used as an additional instrument to assess risk factors and potential prognostic information. In this chapter, we discuss such studies and how the evaluation may benefit from visual analytics. Cluster analysis to define groups, reliable image analysis of organs in medical imaging data and shape space exploration to characterize anatomical shapes are among the visual analytics tools that may enable epidemiologists to fully exploit the potential of their huge and complex data. To gain acceptance, visual analytics tools need to complement more classical epidemiologic tools, primarily hypothesis-driven statistical analysis.
Bernhard Preim, Paul Klemm, Helwig Hauser, Katrin Hegenscheid, Steffen Oeltze, Klaus Toennies, Henry Völzke

Three Dimensional Visualisation of Microscope Imaging to Improve Understanding of Human Embryo Development

The analysis of processes on a cellular and sub-cellular level plays a crucial role in life sciences. Commonly microscopic assays make use of stains and cellular markers in order to enhance image contrast, but in many cases, cell imaging requires the sample to be undisturbed during the imaging process, making staining, dying and fixing impractical. Non-destructive techniques are especially useful in long term imaging or in the study of sensitive cell types, such as stem cells, embryos or nerve cells. Novel advances in computation, imaging and incubator technology have recently made it possible to prolong the imaging time, reduced the cost of storing data and opened a door to the development of new computer aided analytical tools based on microscopic image data. Here we illustrate how Hoffman Modulation Contrast imaging and Confocal Microscopy can be combined with visual computing and present results from determination of cell number, volume, spatial location and blastomere connectivity, using examples from embryos grown for in vitro fertilisation. We give examples of how knowledge of the imaging technique can be used to further improve the computer analysis and also how visually guided tools may aid in the diagnostic interpretation of image data and improve the result. Finally we discuss how the use of microscopic data as a basis for embryo modelling may help in both research and educational purposes. The aim of this chapter is to give an example of how microscopic imaging can be combined with standard computer vision techniques to aid in the interpretation of microscopic data, and demonstrate how visual computing techniques can make an essential difference in terms of scientific output and understanding.
Anna Leida Mölder, Sarah Drury, Nicholas Costen, Geraldine Hartshorne, Silvester Czanner

Quantitative Analysis of Knee Movement Patterns Through Comparative Visualization

In this paper, we present a novel visualization approach for the quantitative analysis of knee movement patterns in time-varying data sets. The presented approach has been developed for the analysis of patellofemoral instability, which is a common knee problem, caused by the abnormal movement of the patella (kneecap). Manual kinematic parameter calculations across time steps in a dynamic volumetric data set are time-consuming and prone to errors as well as inconsistencies. To overcome these limitations, the proposed approach supports automatic tracking of identified features of interest (FOIs) in the time domain and, thus, facilitates quantitative analysis processes in a semiautomatic manner. Moreover, it allows us to visualize the movement of the patella in the femoral groove during an active flexion and extension movement, which is essential to assess kinematics with respect to knee flexions. To further support quantitative analysis, we propose kinematic plots and time-angle profiles, which enable comparative dynamics visualization. As a result, our proposed visualization approach facilitates better understanding of the effects of surgical interventions by quantifying and comparing the dynamics before and after the operations. We demonstrate our approach using clinical time-varying patellofemoral data, discuss its benefits with respect to quantification as well as medical reporting, and describe how to generalize it to other complex joint movements.
Khoa Tan Nguyen, Håkan Gauffin, Anders Ynnerman, Timo Ropinski

Visualization in Life Sciences


Interactive Similarity Analysis and Error Detection in Large Tree Collections

Automatic feature tracking is widely used for the analysis of time-dependent data. If the features exhibit splitting behavior, it is best characterized by tree-like tracks. For a large number of time steps, each with numerous features, these data become increasingly difficult to analyze. In this paper, we focus on the problem of comparing and contrasting hundreds to thousands of trees to support developmental biologists in their study of cell division patterns in embryos. To this end, we propose a new visual analytics method called structure map. This two-dimensional, color-coded map arranges trees into tiles along a Hilbert curve, preserving a tree similarity measure, which we define via graph Laplacians. The structure map supports both global and local analysis based on user-selected tree descriptors. It helps analysts identify similar trees, observe clustering and sizes of clusters within the forest, and detect outliers in a compact and uniform representation. We apply the structure map for analyzing 3D cell tracking from two periods of zebrafish embryogenesis: blastulation to early epiboly and tailbud extension. In both cases, we show how the structure map supported biologists to find systematic differences in the data set as well as detect erroneous cell behaviors.
Jens Fangerau, Burkhard Höckendorf, Bastian Rieck, Christian Heine, Joachim Wittbrodt, Heike Leitte

Efficient Reordering of Parallel Coordinates and Its Application to Multidimensional Biological Data Visualization

Multidimensional data visualization is a challenging research field with many applications in various fields of sciences. Parallel coordinate plots are one of the most common information visualization techniques for visualizing multidimensional data. Unfortunately, the effectiveness of parallel coordinates depends heavily on the order of the data dimensions and different orders exhibit different information about the structures in the multidimensional data. In this paper, we propose a method that supports an automatic dimension reordering and spacing of the axes in parallel coordinate plots. The underlying idea of our method is to find an asymptotic for the optimization of the permutation based on data dimension similarity. We present our method with two kinds of similarities, namely, Pearson’s correlation similarity for unclassified data and class distance consistency for classified data. We present results on well-known multidimensional data sets to show how our method improves the parallel coordinate plots and to prove its efficiency. Finally, we demonstrate how our approach can be applied to the visualization of bivariate structures in biological data.
Tran Van Long, Lars Linsen

Extraction of Robust Voids and Pockets in Proteins

Voids and pockets in a protein, collectively called as cavities, refer to empty spaces that are enclosed by the protein molecule. Existing methods to compute, measure, and visualize the cavities in a protein molecule are sensitive to inaccuracies in the empirically determined atomic radii. This paper presents a topological framework that enables robust computation and visualization of these structures. Given a fixed set of atoms, cavities are represented as subsets of the weighted Delaunay triangulation of atom centres. A novel notion of \((\varepsilon,\pi )\)-stable cavities helps identify cavities that are stable even after perturbing the atom radii by a small value. An efficient method is described to compute these stable cavities for a given input pair of values \((\varepsilon,\pi )\). This approach is used to identify potential pockets and channels in protein structures.
Raghavendra Sridharamurthy, Talha Bin Masood, Harish Doraiswamy, Siddharth Patel, Raghavan Varadarajan, Vijay Natarajan


Additional information

Premium Partner

    Image Credits