Since the invention of x-ray CT, computational techniques have played an important role in biomedical imaging. Now, most modern imaging systems acquire coded data and use integral equations of the first kind to code spatial and/or spectral information.
Such a computational approach is essential for noninvasive acquisition of biomedical images that reveal the structures, metabolism, and function of internal tissues or organs in humans and/or other biological systems. With the rapid development of high-performance computing and machine learning technology, we are entering a new era of computational imaging, in which advanced computational methods can fundamentally transform how biomedical images are acquired and processed. This working group, with expertise in image data acquisition, image reconstruction, image processing, machine learning, and computational modeling as well as clinical applications, is focused on developing advanced computing theory, algorithms, and software tools for next-generation imaging systems. We aim to solve fundamental computational and algorithmic problems to enable a new imaging paradigm that will effectively integrate physics modeling, biological modeling and machine learning into the imaging process to overcome many of the existing technical barriers in imaging speed, spatial resolution, and signal-to-noise ratio. Our research work is expected to have direct scientific and clinical impacts. For example, in brain imaging, the proposed technical advances will allow neuroscientists and doctors to acquire neuroimaging data with much higher accuracy and efficiency and richer information than any existing methods.