Computational Scattered Light Imaging for Label-free Deep Living Systems 

Abstract: Imaging deep inside living organisms with cellular resolution remains challenging as recurrent scattering by submicron-scale heterogeneous structures degrades the image-bearing capability of light. In this talk, I will present my work on imaging through turbid tissues with improved resolution and sensitivity using interferometric techniques and computational algorithms. I will first discuss how quantitative phase imaging (QPI) improves contrast and sensitivity in three-dimensional cellular structures. I will show how the phase information provided by QPI is used to retrieve the intrinsic properties of biological samples. Then, I will present a label-free imaging method using deep learning to achieve confocal-level resolution, sensitivity, and chemical specificity non-destructively on unlabeled specimens. Next, I will present an ongoing project on confocal gradient light interference microscopy for in vivo mouse brain imaging with a comparison to multiphoton imaging. The talk will conclude with a discussion of my future research direction on pushing the limits of label-free optical imaging in deep living systems and artificial intelligence-assisted applications.