Lighthouse projects

MedViz Lighthouse Project – Multimodal imaging and ultrasound microbubble drug delivery in targeted cancer therapy, managed by Professor and Chief physician Bjørn Tore Gjertsen, Clinical Research unit, Section for research and innovation, Haukeland University Hospital and Clinical Department K2, University of Bergen

Project background

The project focuses on the development of sonoporation as a novel treatment modality for cancer, but also searches for other disease were this modality is feasible. Ultrasound has a role in the delivery of therapeutic agents, including genetic material, protein and chemotherapeutic agents. Cavitating gas bodies, such as micro-bubbles, are the mediators through which the energy of relatively non-interactive pressure waves is concentrated to produce forces that permeabilise cell membranes and disrupt the vesicles that carry drugs. Thus, the presence of micro-bubbles greatly enhances ultrasonic delivery of bioactive substances. You may read more about this project in the MedViz Annual Report from 2015.


MedViz Lighthouse Project – Image-based quantitative assessment on abdominal organ function, managed by Professor and Chief physician Jarle T. Rørvik, Department of Radiology, Haukeland University Hospital and Clinical Department K1, University of Bergen

Project background

Recent advances in medical technology have shown the great potential of imaging modalities like MRI, PET, SPECT, US, and optical imaging to provide important functional information about tissues and organs, and thus go beyond morphological characterization only. This is the case for abdominal organs as well. However, abdominal organs are particularly challenging due to movements during the imaging session, where the organ displacements are caused by respiration, pulsations, and peristalsis. In contrast to functional imaging of the brain, being a rather stationary organ, dynamic contrast enhanced MRI (DCE-MRI), blood-oxygenation level dependent MRI (BOLD-MRI), diffusion-weighted MR imaging (DW-MRI) and arterial spin labelling imaging (ASL) of moving organs, like the kidney, have been lagging behind with respect to their implementation, quantification and clinical applicability. In order to improve the usefulness of the advanced imaging infrastructure that now exists at Haukeland University Hospital and UoB to also include functional (non-brain) imaging of patients with abdominal disease, further research and technological developments are needed, being based on the standard equipment that is in place. To achieve this goal, we need a new kind of translational research – “from mathematics to medicine”. This consist of a close collaboration and a certain level of mutual understanding between researchers from the basic sciences (mathematics, physics), from informatics and computer science (advanced visualization), from biomedicine (physiology, pharmacokinetics), and medical doctors, i.e. radiologists and clinicians responsible for the relevant diseases and patient groups. During the past few years, our research group (as part of the MedViz consortium) has worked hard to establish an environment for such translational imaging research focusing on kidney diseases. We have carried out and published research on image registration, image segmentation, pharmacokinetic modelling, and visualization. We have organized seminars and meetings (local, and within the EU COST actions), and established cooperation with national and international research groups. We aim to further develop this broad team of competences the coming years by defining specific research goals and subprojects and bring the new knowledge to a broader spectrum of clinical applications, targeting patients with diseases affecting abdominal organs, including the kidney.You may read more about this project in the MedViz Annual Report from 2015.



MedViz Lighthouse Project – Model-supported data visualization to improve the diagnosis process, managed by Professor Helwig Hauser, The Visualization Group, Department of Informatics, University of Bergen

Project Background

Ultrasonography (US) belongs to the most promising live examination modalities, predominantly for its comparably low costs, its patient friendliness and its high speed of acquisition. On the other hand, the interpretation of US images is still challenging, which also makes the navigation by the examiner not fully straightforward.  Back in 2009, we, the Bergen VisGroup in cooperation with national and international partners began the project called IllustraSound,. The main goal was to improve the medical communication based on US images by means of advanced visualization technology.  Within Illustrasound several high-quality results were achieved, and also the project was awarded with the prestigious Dirk Bartz prize for visual computing in medicine on an international level. Moreover, after new praisal by a Nordic evaluation panel in 2013, IllustraSound became one of the three MedViz lighthouse projects, in order to further improve the doctor-to-doctor and the doctor-to-patient communication by means of advanced visualization techniques.  We sketch IllustraSound as the “IllustraSound Tricorder”, which can be represented as a comprehensive next-stage diagnostic tool that brings a high-quality modern health care closer to the patient. Such a tool can provide anatomical and physiological information in real-time. In order to get closer to this goal, major research questions need to be studied.  These questions led us to the following goals:

  1. To further improve the communication between doctors, in particular in the context of telemedicine. We aim, for example, at improved learning of US-novices with the (tele-)support of US-experts.
  2. To achieve the useful fusion of different data types (spatial and non-spatial) and also to integrate different modalities into US-visualization and an US-based exploration pipeline.
  3. To enable quantitative visualization of physiology, based on US imaging.

In-Situ Adaptive Filtering (ISADAF) is a spin-off project of IllustraSound. This project builds on a result from the IllustraSound project: lowest variance streamlines filtering (LVSF) of ultrasound volumes. In the ISADAF project, the researchers from UiB, CMR and HUS jointly developed an acceleration method that allows LVSF. Moreover, the new acceleration technique is versatile and can be applied to other types of filtering methods and therefore has a bigger potential. In September 2014, we published an article about this acceleration method called “visibility-driven processing of streamed volume data”, which particularly addresses streamed 3D ultrasound and the work was also presented at the Eurographics Workshop on Visual Computing for Biology and Medicine in Vienna, Austria. The publication was preceded by a submission of a patent application together with Bergen Teknologi Overføring AS (BTO). Our work was awarded with the NVIDIA best paper award, and we are currently working on an extended version, which we were invited to submit to the high-quality journal Computer Graphics Forum. You may read more about this project in the MedViz Annual Report from 2015.