Adaptive Optics in Microscopy: Practical Use of Wavefront Correctors and Control Methods

Adaptive optics can be used to correct for instrument and sample aberrations or to dynamically shape the wavefront and point spread function (PSF) to create new possibilities or reduce instrument complexity by, for example, reducing the number of mirrors or using cheaper optical components.

This presentation explains how to select the appropriate wavefront corrector and how to integrate it into the hardware or software environment while avoiding common mistakes. Charton also discusses different control strategies. The classical approach is to use a reference point source and a wavefront sensor in closed-loop operation but metric-based control using classical optimization methods such as gradient descent or genetic algorithm can sometimes provide better performances at a lower cost and complexity.

Reinforcement learning using deep neural networks can also extract more significant control signals from the raw data and drive all instrument degrees of freedom, such as parameters, in addition to the wavefront corrector itself.

About the presenter
Julien Charton contributed to the NACO and SPHERE adaptive optics systems for the European Very Large Telescope in Chile. At this time, it was clear that smaller and better wavefront correctors were needed to continue the development of adaptive optics in astronomy and to address emerging new applications such as ophthalmology and microscopy. He started a research and technology project in the IPAG laboratory in Grenoble, France, to address this issue, soon leading to a new kind of electromagnetic deformable mirror. He co-founded the ALPAO company to develop and commercialize this new technology and is currently its chief scientific officer. ALPAO now employs 40 people and is a leading provider of adaptive optics components and systems.