User Publications

This pages lists documented users of Voreen as well as references to Voreen in a number of categories.
Please contact us in case your publication is not listed here.


Scientific Publications

Automated Segmentation of Immunostained Cell Nuclei in 3D Ultramicroscopy Images
Aaron Scherzinger, Florian Kleene, Cathrin Dierkes, Friedemann Kiefer, Klaus H. Hinrichs, Xiaoyi Jiang
Pattern Recognition - 38th German Conference, GCPR 2016, Hannover, Germany, September 12-15, 2016, Proceedings, Lecture Notes in Computer Science, Volume 9796, page 105-116 - 2016

"The images in this paper have been rendered using the framework Voreen."

Fossilized digestive systems in 23 million-year-old wood-boring bivalves
Steffen Kiel, Stefan Götz, Enric Pascual-Cebrian, Dominik K. Hennhöfer (University of Göttingen / Heidelberg University)
Journal of Molluscan Studies, Volume 78, Issue 4, Pp. 349-356 (2012)

"Digital reconstructions were produced from these subsamples using OsiriX biomedical image software, Meshlab v. 1.3.0a and Voreen v. 2.6.1."

"The Voreen Team is thanked for its visualization software package ("

Imaging of the mouse lung with scanning laser optical tomography (SLOT)
Manuela Kellner, Marko Heidrich, Rebecca Beigel, Raoul-Amadeus Lorbeer, Lars Knudsen, Tammo Ripken, Alexander Heisterkamp, Heiko Meyer, Mark Philipp Kühnel, Matthias Ochs (Hannover Medical School / Laser Zentrum Hannover / REBIRTH Cluster of Excellence / German Center for Lung Research / University Jena)
Journal of Applied Physiology vol. 113 no. 6 975-983

"Rendering was performed using open source Volume Rendering Engine (Voreen)"

Design and Evaluation of Multifield Visualisation Techniques for 2D Vector Fields
Chris C. van Egmond, Charl P. Botha, Burkhard C. Wünsche (University of Auckland / TU Delft)
IVCNZ '12 Proceedings of the 27th Conference on Image and Vision Computing (2012)

"We created LIC textures of vector fields using Voreen [16] and used concepts from graphics and cognitive psychology to combine them. Except for the bump map, all vector fields were rendered using an implementation of the fast LIC algorithm [22] in Voreen (sampling each pixel and using a large convolution kernel k = 100)."

GPU-Based Volumetric Reconstruction and Rendering of Trees From Multiple Images
Dominik M. M. Vock, Stefan Gumhold, Marcel Spehr, Joachim Staib, Patrick Westfeld, Hans-Gerd Maas
The Photogrammetric Record Volume 27, Issue 138, pages 175–194, June 2012

"The smaller the determined error values are, the more the volume resembles the expected tree model. For better visualisation of the MSE, the error values are projected onto an interval between 0 and 1 and prepared as an independent volume. As the error can strongly vary within different regions of the model, a combined rendering of the error volume and the actual reconstruction is recommended. This is achieved in this paper by using the volume rendering engine Voreen (Voreen, 2010)."

Highly efficient 3D fluorescence microscopy with a scanning laser optical tomograph
Raoul-Amadeus Lorbeer, Marko Heidrich, Christina Lorbeer, Diego F. Ramírez Ojeda, Gerd Bicker, Heiko Meyer, and Alexander Heisterkamp ()
Optics Express, Vol. 19, Issue 6, pp. 5419-5430 (2011)

"These data sets then were clipped with ImageJ and visualized using Voreen."

"We would like to thank [...] the program developers of the free software ImageJ ( and Voreen [...]"

X-ray based methods for 3D characterization of charge collection and homogeneity of sensors with the use of Timepix chip
Zemlicka, J.; Jakubek, J. ;  Jakubek, M.;  Vykydal, Z.;  Chelkov, G.A.;  Kruchonok, V.G.;  Elkin, V.G.;  Fiederle, M.;  Fauler, A.; Tolbanov, O.P.;  Tyazhev, A.V.;  Visser, J.  (Inst. of Exp. & Appl. Phys., CTU in Prague)
Nuclear Science Symposium and Medical Imaging Conference (NSS/MIC), 2011

"Visualization of the data was done in Voreen software [5]."

3D imaging of biofilms on implants by detection of scattered light with a scanning laser optical tomograph
Marko Heidrich, Mark P. Kühnel, Manuela Kellner, Raoul-Amadeus Lorbeer, Tineke Lange, Andreas Winkel, Meike Stiesch, Heiko Meyer, Alexander Heisterkamp(Laser Zentrum Hannover / Hannover Medical School / Universität Hannover, CrossBIT Verbundzentrum für Biokompatibilität und Implantatimmunologie in der Medizintechnik)
Biomed Opt Express. 2011 November 1; 2(11): 2982–2994.

"Rendering was performed using Voreen."

"We would like to thank the program developers of the open source software ImageJ ( and Voreen."

Investigation of Temperature Statistics in Turbulent Rayleigh-Bénard Convection using PDF Methods
J Lülff, M Wilczek, R Friedrich (University of Münster)
Journal of Physics: Conference Series (2011)

"Figure 1: Volume renderings of instantaneous snapshots of two corresponding fields. Both volume renderings have been done with the open source renderer Voreen (Meyer-Spradow et al., 2009)."

Temperature statistics in turbulent Rayleigh–Bénard convection
J Lülff, M Wilczek, R Friedrich (University of Münster)
New Journal of Physics, Volume 13, January 2011

"Figure 1. Snapshot of the temperature field T (r). The green, blue and red arrows
point in the x-, y- and z-directions, respectively. The volume rendering has been
done with the open source renderer Voreen [5]."

Application of Tensor Approximation to Multiscale Volume Feature Representations
Susanne K. Suter, Christoph P. E. Zollikofer, and Renato Pajarola ()
Vision, Modeling, and Visualization (2010)

"Some of the images used in this paper have been generated using the Voreen volume rendering engine ( [MSRMH09]."

Scientific Papers about Voreen-Extensions

Dual-modal visibility metrics for interactive PET-CT visualization
Younhyun Jung, Jinman Kim, David Dagan Feng (University of Sydney)
Engineering in Medicine and Biology Society (EMBC), 2012 Annual International Conference of the IEEE

"As a visual feedback mechanism, dm-vh needs to be computed in interactive volume rendering. To achieve
real-time performance, we used the GPU computing [16]. We used the Voreen volume rendering library [17], an open source engine that allows interactive visualization of volumetric data sets with high flexibility for integrating new visualization techniques."

Accumulation of local maximum intensity for feature enhanced volume rendering
Ronghua Liang, Yunfei Wu, Feng Dong, Gordon Clapworthy (Zhejian University of Technology, Hangzhou / University of Bedfordshire, Luton)
The Visual Computer, Volume 28, Issue 6-8, pp 625-633

"The proposed method was implemented on PC platform with an Intel Core 2 CPU of 2.10 GHz and an NVidia GTX 260 graphic card with 1 G video memory. We use an open-source toolkit—Voreen—for the rendering [25]. It followsthe standard raycasting pipeline. Pre-processing involves a Gaussian filter implemented in a GPU fragment program."

Efficient Volume Rendering on the Face Centered and Body Centered Cubic Grids
Max Morén (Uppsala University)
Student thesis (2012)

"Described in this thesis is the development of an extension for the volume rendering engine Voreen, implementing two recently presented  GPU accelerated reconstruction algorithms for these grids, along with a simple nearest neighbor method."

Removal of Surface Artifacts of Material Volume Data with Defects
Jie Shen, Vela Diego, David Yoon (University of Michigan)
Computational Science and Its Applications - ICCSA 2011, Volume 6783, 2011, pp 624-634

"Furthermore, multiresolution visualization of material defects is implemented as a visual way to inspect the correctness of our LD method. It is based on VOREEN [53], which is an open-source volume rendering engine."

Real-time 4D ultrasound visualization with the Voreen framework
Ralf Bruder, Philipp Jauer, Floris Ernst, Lars Richter, Achim Schweikard (University of Lübeck)
ACM SIGGRAPH 2011 Posters

Voreen is an open source volume rendering engine which allows interactive visualization of volumetric data sets with high flexibility when integrating new visualization techniques [Meyer-Spradow et al. 2009]. It is designed with minimal overhead, so that, even for difficult processing networks, high rendering speeds can be achieved. While Voreen is well-equipped to display Cartesian volumes, volumes in polar coordinates, such as raw ultrasound beam data, currently have to be interpolated and converted to Cartesian coordinates to be displayed. This is highly time-consuming on a CPU while interpolation and volume rendering can be done on a graphics card in real-time [Sumanaweera 2004].

Volume Rendering for Planning and Performing Neurosurgical Interventions
U. Eisenmann, A. Freudling, R. Metzner, M. Hartmann, C. R. Wirtz, H. Dickhaus (University Hospital Heidelberg/Ulm)
IFMBE Proceedings, Volume 25/6, 2009, pp 201-204

"In this paper we analyze two neurosurgical scenarios where volume rendering could improve the planning and/or intervention process. Three different volume rendering libraries (VTK, voreen and VGL) and the surface-based visualization of our Multimodal Operation Planning System (MOPS 3D) were compared according to these scenarios.

The results of the comparison show that volume rendering offers a superior visualization quality.

Despite of up-to-date hardware only one library (voreen) was able to achieve interactive frame rates and high quality output on the test dataset during the evaluation."

Integrated Software for Fusion of CT- and Rotational Angiography for Image Guided Neurosurgery on Cerebral Aneurysms
R. Metzner, J. Rüppell, S. Haase, R. Floca, U. Eisenmann, M. Hartmann, C. R. Wirtz,H. Dickhaus
IFMBE Proceedings Volume 25/6, 2009, pp 263-266

"As an alternative to the VTK filtering procedure a faster hardware based volume rendering of the RA data by voreen is offered."


In Scientific Publications

These are only publications where Voreen has been commented on. For more references to our paper about the Voreen architecture see Google Scholar.

MRI-Based Visualisation and Quantification of Rheumatoid and Psoriatic Arthritis of the Knee
Ben Donlon, Douglas Veale, Patrick Brennan, Robert Gibney, Hamish Carr, Louise Rainford, ChinTeck Ng, Eliza Pontifex, Jonathan McNulty, Oliver FitzGerald (UCD, Ireland)
Visualization in Medicine and Life Sciences, Mathematics and Visualization 2012, pp 45-59

"While it is true that there are many commercial and open source software packages available which allow high quality interactive visualisations of volumetric datasets such as Voreen, SCIRun, AMIRA [23–25], the key benefit of developing a custom designed application is that it can be tailored to suit a specifically defined problem, with a high level of input from the medical professionals that will use it, as to their exact requirements."

Computergrafik und Bildverarbeitung
Alfred Nischwitz, Max Fischer, Peter Haberäcker,Gudrun Socher
2012, pp 72-120

"Darauf aufbauend gibt es ein high-level Werkzeug zum Volume Rendering, das quellcode-offene
Voreen (volume rendering engine,, das auch die neuen Möglichkeiten der
Shader Programmierung ausnützt."

CamiTK: A Modular Framework Integrating Visualization, Image Processing and Biomechanical Modeling
Céline Fouard, Aurélien Deram, Yannick Keraval, Emmanuel Promayon ()
Studies in Mechanobiology, Tissue Engineering and Biomaterials Volume 11, 2012, pp 323-354

"Different types of applications, libraries, and more generally frameworks are today available for scientists, most of them are open-source or based on collaborative projects. The lists given here are in no particular order and are not exhaustive but try to give the most popular non-defunct choices. [...] They can be categorized into different area, depending of their main functionality: data visualization (VTK, Paraview, voreen), medical image analyzing or surgical planning (ITK, Analyze, Medical Imaging Interaction Toolkit, MeVisLab, DeViDE, SciRun, amira, 3D Slicer, GIMIAS, MedINRIA/vtkINRIA3D, CreaTools), [...]."

Cloud-based web-service for health 2.0
Denis Vazhenin (Eyes, Japan)
HCCE '12 Proceedings of the 2012 Joint International Conference on Human-Centered Computer Environments, Pages 240-243

Totally different approach in terms of user experience and volume rendering work-flow was realized in Voreen[3]. The main idea of Voreen is to provide users with environment where it is possible to author and perform interactive visualizations of volumetric data. These visualizations can be assembled in form of networks via rapid prototyping. Each network consist of several processors, where each of them perform specialized task for the entire rendering process. This can be task to apply data over ray-casting, geometry creation and rendering or even image processing. Each process can be combined freely with each other, allowing great amount of flexibility and providing at the same time uniform way to conduct volume rendering tasks.

A Shader Framework for Rapid Prototyping of GPU-Based Volume Rendering
Christian Rieder, Stephan Palmer, Florian Link, Horst K. Hahn (MeVis)
Computer Graphics Forum, Volume 30, Issue 3, pages 1031–1040, June 2011

"Voreen [MSRMH09], based on raycasting, provides a graphical user interface for defining a network of so called processors which together assemble a render pipeline. Predefined functions of a shader library can be edited or substituted by the user, which are included into the final shader."

"Voreen facilitates the research of new interactive visualization techniques for volumetric data sets with high flexibility. Our system is limited to customize shading effects and adjust the configuration of the render core at run-time. Extending the render core with additional, configurable features needs to be done in C++."

"In Voreen, functions of a shader library can be interactively edited and are included in the final shader using defines. "

The VRE Volume rendering engine
Michal Hucko, Michal Vanek, Milos Sramek (Comenius University, Bratislava / Austrian Academy of Sciences, Vienna)
SCCG '10 Proceedings of the 26th Spring Conference on Computer Graphics

"Examples of applications based on the data-flow model include MeVisLab [MeVisLab 2009] (Fig. 2) or SCIRun [SCIRun 2009] of University of Utah. Voreen [Meyer-Spradow et al. 2009], although targeted mainly on visualisation, is also based on the data-flow model (Fig. 3)."

Tuvok, an Architecture for Large Scale Volume Rendering
T. Fogal, J. Krüger (University of Utah / DFKI Saarbrücken)
Vision, Modeling, and Visualization (2010)

"Recently, the VisCG at the Universität Münster developed the Voreen system [MSRMH09], a prototyping environment for volume visualization. The interface provided exposes the underlying data flow network and many visualizations require knowledge as to how they are technically realized, which we found was not suitable for a large segment of our user base."

Mapping High-Fidelity Volume Rendering for Medical Imaging to CPU, GPU and Many-Core Architectures
Mikhail Smelyanskiy, David Holmes, Jatin Chhugani, Alan Larson, Douglas M. Carmean, Dennis Hanson, Pradeep Dubey, Kurt Augustine, Daehyun Kim, Alan Kyker, Victor W. Lee, Anthony D. Nguyen, Larry Seiler, and Richard Robb (Intel Corporation / Mayo Clinic)
IEEE Transactions on Visualization and Computer Graphics (TVCG) - 2009

In contrast, there are several GPU implementations of ray-casting using fragment shaders [9, 15]. Mensmann, et al. compared CUDA and fragment-shader ray-casters, both implemented in Voreen rendering engine (, and found that their CUDA code was from 18% slower to 40% faster [20]. These results are not directly comparable to our CUDA implementation, due to their using a 6-neighbor gradient, a different transfer function, and other differences. We performed a more direct comparison by importing our medical imaging datasets and transfer functions into the Voreen fragment shader, as well as by modifying it to compute 26-neighbor gradients. Table 7 gives the results in frames-per-second for the four data sets that we were able to run through Voreen. ds5 and ds6 did not fit into the graphics card’s texture memory.
We see that the Voreen fragment shader is 1.12x to 1.26x faster than our CUDA implementation, which is comparable to the observations made by Mensmann et al. We believe this is largely due to taking advantage of hardware rasterization to perform empty space skipping [15], which is not an option for our CUDA implementation. As the result, for datasets with many empty voxels (Table 3), such as ds2 and ds4, Voreen achieves speedups of 1.23x and 1.28x, respectively, over our CUDA implementation. For datasets with few empty voxels, such as ds1 and ds3, a significant portion of the execution time is spent in gradient computation. As a result, Voreen achieves smaller speedups of 1.12x and 1.15x, respectively. Overall, we see both CUDA and fragment-shader implementations of ray-casting arecomparable in performance.

Other References

MRIcroGL Homepage

"Voreen is an outstanding open-source GLSL volume rendering package"


Interactive GPU volume raycasting in a clustered graphics environment
Christian Noon, Eric Foo, Eliot Winer (Iowa State University)
Proc. SPIE 8316, Medical Imaging 2012: Image-Guided Procedures, Robotic Interventions, and Modeling, 83161L (February 23, 2012)