Imaging Data Science
A growing number of imaging methods are providing not just simple "pictures", but complex and rich data that include three- and four-dimensional views, spectroscopic information on electronic states, or information on response to external stimuli. In addition, the combination of several of these imaging methods with different signals, contrast, and spatial or time resolution scales has become increasingly important.
I3 addresses the need to develop methods and computational resources for the integration and analysis of multimodal image data. For example, we are leveraging Argonne high-performance computing resources in conjunction with state-of-the-art software to develop new mathematical approaches and representations of imaging data. This becomes especially important when one considers that time-resolved tomographic or hyperspectral data sets of a single sample can approach 100 gigabytes in size.