Gabriel Popescu, William L. Everitt Distinguished Professor of Electrical and Computer Engineering University of Illinois at Urbana-Champaign, will lecture on “Phase imaging with computational specificity (PICS) for biomedical applications”
Quantitative phase imaging (QPI) has gained significant interest, especially in the past decade, because of its ability to study unlabeled cells and tissues. As a result, QPI can extract structure and dynamics information from live cells without photodamage or photobleaching. However, in the absence of labels, QPI cannot identify easily particular structures in the cell, i.e., it lacks specificity. This represents the major limitation of QPI when applied in biomedicine.
Inspired by this prior work, we applied deep learning to QPI data, generated by SLIM and GLIM. Gabriel will present present a new microscopy concept, where the process of retrieving computational specificity is part of the acquisition software, performed in real-time. This idea is demonstrated with various fluorescence tags and operation on live cells as well as tissue pathology. Gabriel will explain how this new type of microscopy can potentially replace some commonly used tags and stains and eliminate the inconveniences associated with phototoxicity and photobleaching. Phase imaging with computational specificity (PICS) has an enormous potential for biomedicine.