Computational imaging enabled by engineered flat-optics/metasurfaces

Fluorescence microscopy is a cornerstone technique for analyzing microorganisms, cells, and molecules, driving major breakthroughs in medicine and pharmacology. However, current microscopy systems are often bulky, expensive, and require regular maintenance. They also face trade-offs between form factor, imaging field of view, and resolution. To broaden access, the development of compact, cost-effective, and reliable optical systems is essential.

Flat-optical technologies, particularly nanoscale metasurfaces, offer a transformative approach to imaging by overcoming traditional limitations in resolution, field of view, and system complexity. These ultrathin components enable precise light control, supporting advanced illumination schemes and efficient data capture. When combined with computational algorithms—especially those powered by machine learning—flat-optics can extract high-resolution details from limited datasets. This synergy between innovative illumination and detection techniques is paving the way for scalable, accessible imaging platforms that surpass conventional microscopes. Continued progress in computational imaging is unlocking the full potential of flat-optics, enabling rapid, high-throughput, and high-content imaging tailored to biomedical needs.

This doctoral research focuses on developing a high-resolution fluorescence imaging framework that leverages structured illumination to redefine lens-free, incoherent imaging platforms based on flat-optics technologies. Currently, lens-free fluorescence imaging systems can achieve lateral resolutions approaching cellular dimensions and offer limited 3D imaging capabilities [1-4]. Much of the field’s advancement has stemmed from innovations in light detection. However, theoretical models for image formation remain largely heuristic, limiting their applicability to multiplexed imaging modalities—such as 3D structured illumination microscopy (SIM)—which require integrated design of both illumination and detection optics. The selected PhD candidate will develop a compatible theoretical model for lens-free detection and incorporate it with the current illumination model to form a multiplexed composite system, enabling scalable, high-resolution fluorescence microscopy.

Spatial omics merges spatial context with omics data to reveal molecular distributions and cellular interactions in tissues. High-resolution, wide-field fluorescence imaging is vital but often hampered by long acquisition times due to reliance on traditional optics such as high-NA objectives. This research aims to streamline experimental setups and speed up data collection for spatial omics. By integrating high-throughput imaging with advanced algorithms, the work enables AI-driven analysis that enhances disease understanding, diagnostics, and the discovery of new treatments.