PhD Defence Daniël Cox | Model-based approaches for enhanced wavefront shaping microscopy

Model-based approaches for enhanced wavefront shaping microscopy

Daniël Cox is a PhD student in the department Biomedical Photonic Imaging. (Co)Promotors are prof.dr.ir. I.M. Vellekoop and prof.dr.ir. W. Steenbergen from the faculty Science & Technology (TNW), University of Twente.

In optical microscopy, scattering forms a fundamental limit to imaging depth in biological tissue. Wavefront shaping transcends this limit by utilizing the wave-like nature of optics, using both scattered and unscattered light to create a focus.

Despite impressive demonstrations of wavefront shaping in the last two decades, it is yet to be adopted in biomedical research. We believe the key issues preventing wide-spread adoption are complexity, required level of expertise, and lack of standardization. In this thesis, we've presented smart software solutions to help transition wavefront shaping from a specialized field of expertise to an accessible autonomous tool for biomedical researchers.

In Chapter 2, we presented a technical introduction to wavefront shaping, as well as a study of practical considerations that impact performance, and methods to address them. In addition to this troubleshooting guide, we've provided open source code for an automated troubleshooter that implements these methods. With this guide and code, we hope to lower the threshold for newcomers in the field of wavefront shaping.

A vital component of wavefront shaping microscopy is the spatial light modulator (SLM). Calibration of the SLM is vital but cumbersome, as this requires taking the SLM out of the microscope. In Chapter 3, we presented a novel automated inline calibration technique specifically designed for SLMs in nonlinear microscopy. Our new technique is able to calibrate phase-only SLMs with high precision, while requiring no hardware other than the microscope itself. This technique turns SLM calibration into an easy, automated routine.

One of the factors known to impact wavefront shaping performance is the illumination profile, as this affects the orthonormality of the basis functions on phase-only hardware. In Chapter 4, we presented a method for generating orthonormal optical fields for smooth, arbitrary illumination profiles. We demonstrated a factor 1.5 increase in performance over a non-orthonormal phase-only basis. This work enables automated design of optimized wavefront shaping basis functions.

In Chapter 5, we pioneered a new approach to aberration correction that relies entirely on a priori 3-D information. With our method, we achieved a 10-fold increase in image contrast, on par with an exhaustive search adaptive optics solution. Our new approach requires no feedback and is well-suited for wavefront shaping applications where the sample geometry is smooth and predictable, but with a limited photon budget. We think that this model-based approach could form a great tool in the study of lumen-based organ-on-a-chip systems.

All of our developed solutions are publicly available as open-source software. This includes OpenWFS, an overarching software library for performing and simulating wavefront shaping experiments.

With these contributions, we have brought wavefront shaping microscopy a significant step closer to becoming a widespread tool in biomedical research.