BFS funding to visualization project

Professor Helwig Hauser is, along with the whole visualization group and the Department of Informatics, very happy about the news that BFS will award 7,5 MNOK over four years for their project Visual Data Science for Large Scale Hypothesis Management in Imaging Biomarker Discovery. The BFS funding will be complemented by 12,5 MNOK (by both our Faculty and our Department) to make it 20 MNOK (total project budget):
We see this in line with our previous efforts to strongly contribute to the Bergen MedVis initiative, says Hauser, having attracted extended international recognition over the past decade.  Further, we are excited that two complementary projects get funding, as well (one by Prof. Arvid Lundervold and one by Prof. Ingfrid Haldorsen), with ample possibilities for mutual inspiration and promising synergies. The extended funding of Bergens Forskningsstiftelse, complemented by substantial financial commitments of both our Faculty and our Department, amount to an invaluable support of our research activities and make it possible that we can project our past successes (not at the least in medical visualization) into new strong contributions to the corresponding and rapidly evolving international state of the art.

The project is one of three beneficiaries of this call, which was specific for projects connected to the new Medical Imaging and Visualization Centre (MMIV) which is under establishment at Haukeland University Hospital, and will open on the 11^th December.

Hauser thinks that the funding will be of great importance in providing solutions for better exploiting medical imaging data. Eventually, this will improve the diagnosis of patient conditions as well as the related treatment planning. Besides this long-term significance in terms of future medical benefits due to our enabling technology research, we also find our project important as an up-to-date contribution to cutting-edge research in data science and in our case: visual data science.

The amount and complexity of currently acquired medical imaging data leads to new challenges in knowledge extraction and decision making. Imaging biomarkers are able to improve diagnostics and treatment planning, based on features that can be computed from the measured data. Imaging biomarker discovery is a highly challenging task and traditionally only a single hypothesis (for a new biomarker) can be examined at a time, making it impossible to explore really many as well as more complex potential imaging biomarkers across multi-aspect data. We propose to research and advance visual data science to improve imaging biomarker discovery through the visual integration of multi-aspect medical data with a new visualization-enabled hypothesis management framework. Through the integration of computational approaches and interactive visual analysis techniques, we plan to reduce the time it takes to discover new imaging biomarkers by allowing structured sets of hypotheses to be examined at the same time. Another related goal is to enable the discovery of more complex imaging biomarkers, across multiple modalities, that potentially are able to more accurately characterize diseases. This should lead to a new form of designing innovative and effective imaging protocols and to the discovery of new imaging biomarkers, improving suboptimal imaging protocols and thus also reducing scanning costs. Our project is a truly interdisciplinary research effort, bringing visualization research and imaging research together in one project, and this perfectly suited for the novel Centre for Medical Imaging and Visualization that has been established in Bergen, Norway, this year.

As for the outcome of the project, Hauser says: primarily, this project targets scientific outcome in the form of publications (we expect about two dozens of publications, many of which in high-quality journals/conferences). This goes in line with the plan to lead two new PhD students to successfully finishing their PhD projects in the context of our project. Further, there will be software prototypes of our new visualization solutions, as well as case studies in the context of (at least) three application areas (gynecologic cancer, multiple sclerosis, cognitive aging).

The funding will mainly be used to finance the human resource of this project. This includes (besides in-kind contributions of the PIs) two post-doctoral researchers (3 years) and two PhD students (4 years). The project is an interdisciplinary research effort, integrating visualization research with imaging research, says Hauser: We stand in with the following faculty members: myself, Stefan Bruckner, and Noeska Smit; for imaging research, Renate Gruner will stand in as Co-PI on behalf of the Department of Physics and Technology. In addition to being an interdisciplinary methodological effort, the project relates to three particular application areas:
- Prof. Ingfrid Haldorsen will be a collaborator in this project, contributing with her expertise on imaging biomarker discovery through advanced imaging in gynecologic cancer.
- Prof. Kjell-Morten Myhr will be a collaborator in neurology, standing in for the planned case studies with respect to multiple sclerosis.
- Prof. Arvid Lundervold will be another collaborator from UiB’s Department of Biomedicine, contributing expertise in medical image processing and pattern recognition, as well as longitudinal imaging data from the cognitive aging project.

Also, Hauser adds, we are lucky to have a strong international network, associated with our project, including the following experts:
- Prof. Bernhard Preim (Head of the Visualization group, Dept. of Simulation and Graphics, Univ. of Magdeburg, Germany), contributing his world-renown expertise in medical visualization and visual analysis of medical data, in particular also cohort study data.
- Associate Prof. Anna Vilanova (Computer Graphics and Visualization Group, Dept. of Informatics, Delft Univ. of Technology, The Netherlands), contributing her expertise in biomedical image analysis and interactive visualization of dimensionality reduction methods.
- Prof. Jos Roerdink (Head of the Scientific Visualization and Computer Graphics group, Johann Bernoulli Institute for Mathematics and Computer Science, Univ. of Groningen, The Netherlands), contributing his outstanding expertise in the computational analysis of biomedical data.
- Prof. Peter Filzmoser (Head of research unit Computational Statistics, Institute of Statistics and Mathematical Methods in Economics, Vienna Univ. of Technology, Austria), contributing his exceptional expertise in on statistical data analysis, in particular robust statistics.
- Associate Prof. Jaakko Peltonen (Head of the Probabilistic Machine Learning research group, Department of Computer Science, Aalto University, Finland), contributing his widely respected expertise in machine learning.

On December 11th and 12th, the Medical Imaging and Visualization Centre will officially open with an opening ceremony and conference. These events will feature interactive demonstrations organized by the VisGroup and others, and Helwig Hauser will be presenting the project at the conference. For further details and registration for the conference, please visit the centre webpage at http://www.mmiv.no. You can also find and follow the centre via Facebook and Twitter.