Current clinical practices for the diagnosis and management of diseases often rely on histopathological exami- nation of tissue via optical microscopy. Brightfield imaging of hematoxylin-eosin (H&E)-stained samples repre- sents the predominant approach for accurate and comprehensive evaluation and diagnosis in clinical histopathol- ogy [1, 2]. Additional techniques for disease characterization involve molecularly specific labeling, and use im- munohistochemical or immunofluorescence techniques for brightfield and fluorescence microscopy, respec- tively. Using the latter, multiple analytes can be examined simultaneously [3, 4]. Unfortunately, the complexity of a fluorescent microscope?s optical design scales with the number of multiplexed fluorescent reporters to visual- ize, thus limiting its clinical utility [5]. Another area of interest is to explore clinically relevant information that may exist at spatial resolutions beyond what can be achieved with conventional microscopes. Typical fluorescence microscopy is generally limited by diffraction to an optical resolution of ~200 nm. Though this resolution enables visualization of large cellular structures, it does not support examination of organelle- and suborganelle-level ultrastructure whose morphological changes can correlate with disease, as seen in neurodegeneration, age, and cancer [6-10]. Recently, optical super-resolution technologies have been introduced that achieve imaging reso- lutions better than 50 nm. However, such technologies depend on complex hardware and are currently too costly to be incorporated into typical clinical pathology budgets. Electron microscopy (EM) systems are also an availa- ble option, and routinely image at resolutions of ~1 nm ? however, these are not widely available and are not well suited for molecular specific imaging [11-14]. Additional issues, including size, cost, limited field-of-view, and complexity of sample-prep protocols have prevented EM from being incorporated into standard clinical work- flow. This project will develop a robust, comparatively simple, and low-cost optical system for molecularly-specific multispectral fluorescence imaging at spatial resolutions of ~70 nm, well beneath the classical 200 nm optical resolution limit. To do so, a framework for computational structured illumination (SI) microscopy will be developed to enable super-resolution using uncalibrated illumination patterns. This framework will be deployed using single- wavelength ultraviolet (UV) excitation, which has demonstrated capabilities for simultaneous excitation of multi- ple fluorescent reporters. Specific innovations in this work include a novel reformulation of SI microscopy that uses computational optimization to robustly increase imaging resolution in the presence of system unknowns and imperfections. Furthermore, because UV-based excitation has wavelengths more than a factor of 2 shorter than the fluorophores? visible emission wavelengths, resolution gains by factors greater than 2 are expected, hence enabling sub-100-nm spatial resolutions. If successful, the aims of this project will combine the benefits of multispectral optical imaging with the advantages of sub-100-nm spatial resolution to create a more informative and less demanding alternative to electron microscopy, with applications across biology and histopathology.

Public Health Relevance

The proposed technology will develop a low-cost and robust optical system for efficient examination and assess- ment of histopathology utilizing multispectral imaging at spatial resolutions of ~70 nm. The application of this technology will enable more informative and less demanding visualization alternatives to electron microscopy, for ready incorporation into clinical workflow.

National Institute of Health (NIH)
National Institute of General Medical Sciences (NIGMS)
Postdoctoral Individual National Research Service Award (F32)
Project #
Application #
Study Section
Special Emphasis Panel (ZRG1)
Program Officer
Sakalian, Michael
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
University of California Berkeley
Engineering (All Types)
Biomed Engr/Col Engr/Engr Sta
United States
Zip Code