Sebastian Kazmarek Præsius, PhD
GPU acceleration for real-time 3D ultrasound imaging
GPU acceleration for real-time 3D ultrasound imaging
Project Title: GPU acceleration for real-time 3D ultrasound imaging
Group: Center for Fast Ultrasound Imaging
Supervisors:Jørgen Arendt Jensen, Bernd Dammann, Hans Henrik Brandenborg Sørensen
Project Description:
Super Resolution Ultrasound Imaging using Erythrocytes (SURE) enables blood vessels to be visualized at a higher resolution than conventional methods. It does not require any contrast agents to be injected, and the images are acquired in around four seconds. However, like most super-resolution pipelines, SURE uses a high frame rate and heavy processing, even more so when moving to 3D imaging. The hypothesis is, therefore, that an efficient graphics processing unit (GPU) implementation will be able to process the SURE images/volumes at the same rate that they are acquired, allowing the results to be displayed immediately.
Perspective:
Short bio
Sebastian's studies have mainly focused on mathematics, deep learning, and computer graphics with applications in health technology. He was introduced to ultrasound imaging by Prof. Muyinatu Bell during his exchange stay at Johns Hopkins University in 2021, and he received his M.Sc. degree in Mathematical Modelling and Computation from the Technical University of Denmark, Lyngby, Denmark, in 2022. His master’s thesis was on Real-Time Super-Resolution Imaging using Erythrocytes, and he is currently pursuing a Ph.D. with the Center for Fast Ultrasound Imaging on the same topic expanded to three dimensions.