World-Class Scientific Visualisation
During the international conference "Supercomputing 21", Elisabeth Mayer along with Salvatore Cielo from the Leibniz Supercomputing Centre (LRZ) and research partners at the University College London and Intel, reached the final of the world's six best scientific visualisations. With the three-dimensional images of blood flow in the human forearm, doctors could plan operations and treatments in advance or better understand bodily functions.
The arteries, shown in grey, supply the arm with fresh blood right up to the fingertips, while the colourful veins carry the deoxygenated blood back towards the heart: anyone who wants to can view this process from a variety of perspectives and literally immerse themselves in the network of veins found in the human forearm. With these images, the specialists at the LRZ in Garching reached the finals in the competition for the six best scientific visualisations during the SC21 supercomputing trade fair in St. Louis/USA: They awarded the specialists at the Leibniz Supercomputing Centre (LRZ) in Garching the prize for the best scientific visualisation: "This is a great success and confirms the efforts of the team and our research partners," says Elisabeth Mayer, a staff member at the LRZ's Centre for Virtual Reality and Visualisation (V2C), who was in charge of the visualisation. "The challenge in such projects is to calculate the necessary data and then visualise it, as it creates huge volumes of data that only a supercomputer can process."
Making body functions visible
For almost ten years, the V2C at the LRZ has been converting research results into impressive images, illustrating not only body parts and organs, but also baroque ceiling paintings or climate scenarios. The work on blood flow was done as part of the international research project CompBioMed, for which around 20 universities, research institutes and supercomputing centres have been developing technologies for the digitisation of medicine and pharmacy since 2016 under the leadership of the physician and computer scientist Prof. Dr. Peter Coveney from the University College London (UCL). The focus is on building a "virtual human", a digital twin of the human being. The Centre of Excellence has taken a big step towards this goal with the award-winning visualisation. "We have already seen some spectacular work and virtual reality applications from the V2C," says Prof. Dr. Dieter Kranzlmüller, Director of the LRZ. " I wholeheartedly congratulate Elisabeth Mayer and the team for this important prize. It once again shows the high quality of our services for science and research."
The basis for the internal view of the forearm is provided by an open Lattice-Boltzmann algorithm called HemeLB, which was developed by CompBioMed and models the blood flow in veins three-dimensionally from measured values and image data. For the forearm, HemeLB first calculated for more than 230 million data points and 64-time units the pressure with which the blood is pumped through the arteries and veins during a heartbeat. Each calculation step produced around 7 gigabytes, a total of around 470 gigabytes of information. These values were processed with the open-source graphics programme OSPRay from the Intel oneAPI Rendering Toolkit. Lifting data from one programme to the next does not always work smoothly on a notebook, but in supercomputing it is usually a challenge. Thus, in addition to the visualisation, various tools, plug-ins and workflows were created with which the blood flow can now be displayed much more easily and quickly in other organs and parts of the body. In collaboration with researchers from CompBioMed, the LRZ team is already working on the Circulus Arterius Cerebri, a vascular ring that supplies the brain with blood. Instead of the use case of the forearm, this generates more than 1500 gigabytes of data per time step.
Media for learning and testing operations
"One advantage of our workflow is that any images and media can be created from it and exported," says visualisation specialist Mayer. "Anything is possible - a graphic, a short video clip, even a cinema film could be made from it or a three-dimensional virtual reality application." This would allow doctors to dive into body parts such as the forearm, where they could better understand and comprehend functions. If the models and visualisations are also calculated with individual patient data, it will be possible in future to plan treatment methods and operations or even try them out in advance. A milestone for medicine, which the SC21 jurors awarded as the best SC Scientific Visualisation & Data Analytics Showcase.