AbstractsPhysics

When imaging with a fluorescence microscope, heterogeneity in the index of refraction of the specimen causes aberrations in the images. Image-based sensorless adaptive optics can be used to correct these aberrations. Previous wavefront sensorless adaptive optics schemes have made use of knowledge about the image formation process of a microscope to derive a cost function for an optimisation routine. In this way it was found that e.g. maximising the overall intensity in an image works well for a confocal or multiphoton microscope. However, the image formation in a superresolution Stimulated Emission Depletion (STED) microscope, where a ring-shaped depletion beam is superimposed on the excited sample to suppress fluorescence, is significantly more complicated. When investigating the effect of individual aberration modes on the point spread function (PSF) it can be seen that certain modes impose a specific type of symmetry on the PSF, which depends upon the azimuthal order of the aberration mode. In this thesis, a new metric is presented based on Fourier azimuthal masking (FAM), which makes use of the knowledge of the present symmetry to suppress specimen contributions and enhance the sensitivity of the wavefront reconstruction. This metric is used in combination with semi-definite programming to derive a set of optimal aberration modes for a STED microscopy simulation, which can be used to achieve wavefront reconstruction.