Design and development of non-linear optical microscopy, data acquisition and analysis techniques

Shane Z Sullivan, Purdue University


The ability to perform in vivo real-time diagnostics in both the research and clinical settings is of interest in the respective communities. Current methods require biopsy followed by histopathology, often requiring complex and time consuming sample preparation (i.e. sectioning, staining). These time consuming steps add several complications including obfuscation of the inherent information of interest through preparation techniques. However, perhaps the most significant shortcoming in these techniques is the inability to observe and quantify dynamic processes in the living system. Second harmonic generation (SHG) and other nonlinear optical beam scanning techniques has seen rapid recent development as a fast, non-destructive and quantitative method for in vivo imaging. The major limitation of SHG microscopy and similar techniques is ultimately the signal-to-noise (S/N) in these imaging techniques. With sufficient S/N real-time, video rate, SHG microscopy can be performed to gather insight into living systems as they evolve over time. In pursuit of this goal, fast data acquisition techniques were developed to improve the S/N and imaging speed in SHG microscopy. Methods for enhancing the S/N were developed in modulated signal configurations using digital -in amplifications techniques in SHG microscopy. Additionally, digital filtering techniques were developed to remove significant noise from the raw signal before producing micrographs to greatly enhance the S/N in situations of high noise and uncertainty. Using novel scanning techniques and model based image reconstruction, SHG microscopy was pushed into the kilohertz imaging regime. Additionally, polarization dependent SHG microscopy methods were developed to provide additional structural information at video-rate speeds.




Simpson, Purdue University.

Subject Area

Physical chemistry

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server