Seminar "Data-Heavy Supercomputing in Biomedicine"

The field of biomedical imaging has kept doubling data rates and increasing modalities over the years. We assume that high-volume high-resolution signals encode groundbreaking answers to scientific questions. However, as the acquired signals are not directly interpretable, the promise of groundbreaking discoveries is hindered by our ability to process and leverage the data deluge in a meaningful way. The next large-scale frameworks in biomedicine must be able to decode high-volume experimental information and transform it into scientific insight that can relate to prior knowledge. In my presentation I will talk about different efforts towards this goal, from large-scale image registration and stitching, to large-scale machine‐learning based image segmentation, morphometrics, and large-scale computational continuum mechanics.