Future Perspectives for Parallel Mesh Generation and Real-Time Image-To-Mesh Conversion for Medical Image Computing

RESCHEDULED

Computational Science is one of the three pillars complementing traditional theoretical and physical experimental studies in science and engineering. Parallel finite element mesh generation is a critical building block of the foundation for this pillar and is becoming even more relevant for a growing number of applications with recent advances in high-end and cloud computing. This increases the demand for parallel mesh generation in industry, but there are few experts to meet the market’s needs. One of the reasons is that parallel mesh generation is a relatively new research area transcending the boundaries of two disciplines: computational geometry and parallel computing. In this talk, first we examine how parallel meshing has changed over the last 15 years and then we look forward in terms of trends in high performance computing that will influence the basic research in this area. We will address some of the new algorithmic and software challenges, like managing deep memory/network hierarchies and power consumption for current and emerging high-end parallel architectures.  In the second part, we will look into getting our basic research into the market. We will focus on health care, since today’s regulatory changes affect tomorrow’s disruption in the health care system and the barrier to entry is high for parallel meshing of measured data (eg. images or clouds of points). We will demonstrate how we leverage our work on real-time image-to-mesh conversion for more accurate, faster and cheaper fusion of medical images. Our current focus is on Image Guided Diagnosis and Therapy for brain cancer. The same technology can be customized for Deep Brain Stimulation for Parkinson’s disease as well as prostate and liver cancer.