IEEE VIS 2020: Medical Visualization Highlights

Posted on October 27, 2020 by Noeska Smit

(We are very thankful that Eric Mörth, PhD candidate at the University of Bergen VisGroup and MMIV, could write this report looking forward to the medical visualization-related papers at IEEE VIS 2020 for us!)

It is this time of the year again. The biggest visualization conference is happening again, and we are all very excited about it. This year VIS is completely virtual which comes with its challenges! For me, one positive side of it is the new VIS2020 virtual webpage. All papers are nicely presented, and it is quite easy to find the papers you are interested in. Nevertheless, I still have the feeling that I would miss something if I don’t go through all the papers and mark the papers which are about medical visualization. This year I thought instead of making a schedule of all medical visualization papers only for myself, I am sharing my schedule with you, so you can enjoy all the great works of the authors as well! I hope to see you all in the Discord channels discussing the awesome papers!

VIS Short Papers

Best Papers:

The Anatomical Edutainer

Marwin Schindler, Hsiang-Yun Wu, Renata Georgia Raidou

View paper page – View presentation: Tuesday, October 27th, 2020 @ 16:50 GMT+01:00

The authors present a workflow called Anatomical Edutainer, which enables the easy, accessible, and affordable generation of physicalizing for tangible, interactive anatomical edutainment. They use 2D printable and 3D foldable physicalizations which change their visual properties under colored lenses or colored lights.

Visualizing Machine Learning:

Explainable Spatial Clustering: Leveraging Spatial Data in Radiation Oncology

Andrew Wentzel, Guadalupe Canahuate, Lisanne van Dijk, Abdallah Mohamed, Clifton David Fuller, G. Elisabeta Marai

View paper page – View presentation: Friday, October 30th, 2020 @ 16:00 GMT+01:00

The authors deliver a set of lessons learned for creating visual and explainable spatial clustering for clinical users. Their insights were gathered from multi-years collaboration with radiation oncologists and statisticians.

Vis and Scientific Computing

PRAGMA: Interactively Constructing Functional Brain Parcellations

Roza Gunes Bayrak, Nhung Hoang, Colin Blake Hansen, Catie Chang, Matthew Berger

View paper page – View presentation: Tuesday, October 27th, 2020 @ 20:20 GMT+01:00

The authors present PRAGMA, an interactive visualization method that allows domain experts to derive scan-specific parcellations from established atlases. PRAGMA features a hierarchical clustering scheme for defining temporally correlated parcels in varying granularity. The visualization supports the user in deciding on how to perform the clustering. The authors assessed the effectiveness of PRAGMA through a user study with four neuroimaging domain experts.

VIS Full Papers:

Health & Disease

Humane Visual AI: Telling the stories behind a medical condition

Wonyoung So, Edyta Paulina Bogucka, Sanja Scepanovic, Sagar Joglekar, Ke Zhou, Daniele Quercia

View paper page – View presentation: Friday, October 30th, 2020 @ 15:00 GMT+01:00

The authors mined and combined information from around half a million Reddit posts and open prescription data from the National Health Service in England to visually communicate each of the 14 selected medical condition’s biological, psychological and social aspects through storytelling. A user study with 52 participants delivered interesting insights about the effect of the visualization on them

QualDash: Adaptable Generation of Visualisation Dashboards for Healthcare Quality Improvement

Mai Elshehaly, Rebecca Randell, Matthew Brehmer, Lynn McVey, Natasha Alvarado, Chris P. Gale, Roy Ruddle

View paper page – View presentation: Friday, October 30th, 2020 @ 15:15 GMT+01:00

The authors present a task analysis that resulted in a metric card metaphor as a unit of visual analysis in healthcare quality improvement. They are using the concept as a building block for generating highly adaptive dashboards and leading to the design of a metric specification structure. QualDash has been deployed in cardiology wards and pediatric intensive care units in five NHS hospitals and the authors report on evaluation results of the usage in a real-world scenario.

Visualization of Human Spine Biomechanics for Spinal Surgery

Pepe Eulzer, Sabine Bauer, Francis Kilian, Kai Lawonn

View paper page – View presentation: Friday, October 30th, 2020 @ 15:30 GMT+01:00

The authors present a visualization application, designed for the exploration of human spine simulation data. The link simulation outcomes with patient-specific anatomy, to make relevant parameters graspable for clinicians, by introducing new concepts to show the directions of impact force vectors. The authors evaluated their application with both surgeons and biomechanical researchers.

In Search of Patient Zero: Visual Analytics of Pathogen Transmission Pathways in Hospitals

Tom Baumgartl, Markus Petzold, Marcel Wunderlich, Markus Höhn, Daniel Archambault, Michael Lieser, Alexander Dalpke, Simone Scheithauer, Michael Marschollek, Vanessa Eichel, Nico T. Mutters, Tatiana von Landesberger

View paper page – View presentation: Friday, October 30th, 2020 @ 15:45 GMT+01:00

The authors present a novel visual analytics approach to support the analysis of transmission pathways, patient contacts, the progression of the outbreak, and patient timelines during hospitalization. In a final study, feedback from twenty-five experts from seven German hospitals provided evidence that their solution brings significant benefits for analyzing pathogen outbreaks.

DPVis: Visual Analytics with Hidden Markov Models for Disease Progression Pathways

Bum Chul Kwon, Vibha Anand, Kristen A Severson, Soumya Ghosh, Zhaonan Sun, Brigitte I Frohnert, Markus Lundgren, Kenney Ng

View paper page – View presentation: Friday, October 30th, 2020 @ 16:00 GMT+01:00

The authors introduce DPVis to integrate model parameters and outcomes of Hidden Markov Models into an interpretable and interactive visualization. The authors state that their tool is successful in evaluating disease progression models, visually summarizing disease states, interactively exploring disease progression patterns, and building, analyzing, and comparing clinically relevant patient subgroups.

Molecules, Cells & Vessels

Visual cohort comparison for spatial single-cell omics-data

Antonios Somarakis, Marieke Ijsselsteijn, Sietse Luk, Boyd Kenkhuis, Noel de Miranda, Boudewijn Lelieveldt, Thomas Höllt

View paper page – View presentation: Wednesday, October 28th, 2020 @ 17:15 GMT+01:00

The authors present an interactive visual analysis workflow for the comparison of cohorts of spatially resolved omics-data. They allow for a comparative analysis of two cohorts based on multiple levels-of-details. The application enables the identification of cohort-differentiating features and outlier samples at any stage of the workflow. The authors conducted multiple case studies with domain experts from different application areas and with different data modalities, to show the effectiveness of the workflow.

Improving the Usability of Virtual Reality Neuron Tracing with Topological Elements

Torin McDonald, Will Usher, Nate Morrical, Attila Gyulassy, Steve Petruzza, Frederick Federer, Alessandra Angelucci, Valerio Pascucci

View paper page – View presentation: Wednesday, October 28th, 2020 @ 17:30 GMT+01:00

The authors propose a new semi-automatic method to guide users in tracing neurons by using topological features. The use of a virtual reality framework which has been used for manual tracing before. In a pilot study, neuroscientists demonstrated a strong preference for their tool over prior approaches. The approach delivered an increased tracing speed while retaining a similar accuracy compared to a fully manual approach.

CMed: Crowd Analytics for Medical Imaging Data

Ji Hwan Park, Saad Nadeem, Saeed Boorboor, Joseph Marino, Arie Kaufman

View paper page – View presentation: Wednesday, October 28th, 2020 @ 17:45 GMT+01:00

The authors present CMed, a visual analytics framework for the exploration of medical image data annotations, acquired from crowdsourcing. They evaluated the efficacy of the framework with two medical crowdsourcing studies and provide expert’s feedback to show the effectiveness of CMed.

Void Space Surfaces to Convey Depth in Vessel Visualizations

Julian Kreiser, Pedro Hermosilla, Timo Ropinski

View paper page – View presentation: Wednesday, October 28th, 2020 @ 18:00 GMT+01:00

The authors propose Void Space Surfaces, a technique that utilizes the empty space in between vessel branches to communicate the depth and their relative positioning. It allows them to improve the depth perception of the vascular structures without interference with the spatial data. Furthermore, the authors deliver two user studies to evaluate the perceptual impact of Void Space Surfaces.

CG&A Full Papers

Towards Better Visual Analysis

Towards Placental Surface Vasculature Exploration in Virtual Reality

Johannes Novotny, Wesley R. Miller, François I. Luks, Derek Merck, Scott Collins, David H. Laidlaw

View paper page – View presentation: Thursday, October 29th, 2020 @ 18:15 GMT+01:00

The authors present a case study where they evaluation the application of virtual reality environments to identify placental surface blood vessels. They observed that the visualization is easy to understand and allows for intuitive exploration, but complex user interactions remained a challenge.

Schedule:

You can also download this schedule with session links: PaperScheduleMedVis Download

I am looking forward to a successful virtual version of our biggest and most impactful conference, VIS. I hope to see you in one of the Discord channels to discuss all the mentioned papers and to get the feeling of being a part of this year’s VIS conference. Let us hope that future conferences will be in person again to enjoy the experience even more!

See you at VIS2020 😊

Medical Visualization PhD defenses

Posted on February 15, 2015 by Noeska Smit

I apologize for the delay in reporting this, but I heard about/attended three great medical visualization PhD defenses late 2014:

Dr. Mathias Goldau successfully defended his PhD at the University of Leipzig. His thesis is entitled ‘Multi-modal and Slice-based Visualizations of Diffusion Tractography Data’ and you may or may not already be familiar with his work on OpenWalnut, an open source tool for multi-modal medical and brain data visualization. The full text of the thesis can be downloaded here or take a look at his publications here.
Dr. Peter Kok successfully defended his PhD at the Technical University of Delft. In his thesis, ‘Integrative Visualization of Whole Body Molecular Imaging Data‘, Dr. Kok present methods to map molecular imaging data to a common reference frame, to combine multiple modalities and to compare scans taken at different timepoints. The full text of the thesis is available here.
Dr. Kok with his paranymphs and the thesis committee members. tblr: Jos Roerdink, Elmar Eisemann, Charl Botha, Erik Jansen, Boudewijn Lelieveldt, beadle Rina Abbriata, Louise van der Weerd, Bernhard Preim, Noeska Smit, Peter Kok, Thomas Kroes and the head of the committee.
On the same day, Dr. Stef Busking successfully defended his PhD at the Technical University of Delft as well. His thesis, ‘Visualization of Variation and Variability‘ deals with comparative visualization as a means to analyze variation or variability based on two or more specific instances of the data. The full text of the thesis is available here.

Congratulations to all three doctors!

PhD Thesis Rocco Gasteiger defended with summa cum laude

Posted on March 26, 2014 by Noeska Smit

(This news was submitted by the Magedeburg Visualization group. We would like to congratulate Rocco Gasteiger with his outstanding work and his summa cum laude defense!)

Rocco Gasteiger

On February 7, 2014 Rocco Gasteiger successfully defended his PhD Thesis “Visual Exploration of Cardiovascular Hemodynamics” with summa cum laude in the Visualization Group (headed by Prof. Bernhard Preim) at the Otto-von-Guericke University (OVGU) Magdeburg. During the work on his PhD Thesis, Rocco provided new contributions to the field of visual exploration of complex morphological and hemodynamic data of cardiovascular diseases with a focus on cerebral aneurysms. In the current clinical and biomedical research, the analysis of measured and simulated blood flow data is of major importance for the examination of diseases, their courses and optimized patient-specific treatment options. The main contributions of the PhD Thesis consist of the development of an expressive visualization method for vessel illustrations with embedded flow information, the automatic extraction of qualitative hemodynamic parameters (see Fig. 1) as well as a flexible focus and context illustration for the examination of multi-parameter data of the hemodynamics (also known as FlowLens).

Automatic extracted and illustratively depicted Visualization of the inflow jet (arrow glyph) and impingement zone (contours) of the blood flow in a cerebral aneurysm.

The articles were published in the well-known Journal IEEE Transactions on Visualization and Computer Graphics (TVCG) and the Eurographics Workshop on Visual Computing in Biology and Medicine (VCBM). The articles resulted from a close cooperation with his colleague Mathias Neugebauer, the Visual Computing group at OVGU (headed by Prof. Holger Theisel) as well as Prof. Anna Vilanova Bartroli (TU Delft) and Dr. Roy van Pelt (TU Eindhoven). Besides the work for his PhD Thesis, Rocco contributed to further topic-related publications as coauthor. Amongst others, his work was recognized by the medvis award (1st price in 2012). This award is offered by the Medical Visualization group of the German Informatics Society (GI) and is sponsored by BrainLab. Furthermore, the joint works of Mathias, Rocco, Gábor Janiga (Institute of Fluid Dynamics and Thermodynamics, OVGU), Oliver Beuing (University Hospital Magdeburg) and Prof. Bernhard Preim received the 2nd Dirk Bartz Prize for Visual Computing in Medicine at Eurographics 2013.

Groundbreaking volume rendering papers by Professor Karl Heinz Höhne: a medvis.org exclusive!

Posted on January 30, 2012 by Noeska Smit

As medvis professionals, we are accustomed to volume rendering as an every-day tool for exploration of medical data. The two technical papers below represent one of the first milestones in rendering volumetric data. We are extremely happy and excited that Professor Karl Heinz Höhne allowed us to post these papers on our blog. Published in 1986 and 1988, these papers cannot be found online anywhere else and proposed groundbreaking volume rendering techniques.

Combined 3D-display of conventional and angiographics MR data.

The first paper entitled ‘Shading 3D-images from CT Using Gray-Level gradients’ by Karl Heinz Höhne and Ralph Bernstein was published in 1986 in the IEEE Transactions on Medical Imaging. This paper describes a shading method based on the partial volume effect using the gray-level gradients along a surface reconstruction of CT images. What sets this method apart from current gradient vector computations is that the gray-values are sampled in screen space and not in voxel space. As a result, surfaces perpendicular to the viewer are brighter than surfaces oriented away from the viewer. In essence, this shading method simulates a headlight configuration.

The second paper entitled ‘Display of Multiple 3D-Objects Using the Generalized Voxel-Model’ by Karl Heinz Höhne, Michael Bomans, Ulf Tiede and Martin Riemer was published in 1988 in the SPIE proceedings. In this paper a generalized voxel-model for the generation of 3D-views from MR-volume data is proposed.

Display of Multiple 3D-Objects Using the Generalized Voxel-Model – hohne88

Using the software and hardware that was available at the time, these papers represent many contributions to the field, for instance selective volume clipping and multimodal visualization. With their work, they created a solid foundation for state-of-the-art (medical) volume rendering techniques. The two papers usually credited as first presenting the ideas behind raycasting, both published in 1988, are:

R. A. Drebin, L. Carpenter, and P. Hanrahan, “Volume rendering,” SIGGRAPH Comput. Graph., vol. 22, no. 4, pp. 65–74, Aug. 1988.
M. Levoy, “Display of surfaces from volume data,” IEEE Computer Graphics and Applications, vol. 8, no. 3, pp. 29-37, May 1988.

Virtual colonscopy best method for screening

Posted on April 25, 2007 by cpbotha

Virtual Colonoscopy with an optical colonoscopy followup, where non-diminutive lesions (< 6mm) are not reported, has been found to be the most cost-effective and safest method for screening. See this MedGadget blog, and this abstract in Cancer for more details.

The TU Delft Visualisation Group is also actively performing research on Virtual Colonoscopy. Our current focus is on automatic polyp detection by using vector field visualisation techniques. Click here for the research project page, or read “L. Zhao, C. P. Botha, J. O. Bescos, R. Truyen, F. M. Vos, and F. H. Post, Lines of curvature for improved diagnosis in virtual colonoscopy, IEEE Transactions on Visualization and Computer Graphics, 2006″ (you can find this article on the group publications page).

Update:
Gerwin has just notified us of a new WebMD article also on this virtual colonoscopy study, this time concerning a new article published in the October 4 2007 issue of the NEMJ.