Smart microscope: an adaptive optics learning system for aberration correction in multiphoton confocal microscopy

Off-axis aberrations in a beam-scanning multiphoton confocal microscope are corrected with a deformable mirror. The optimal mirror shape for each pixe l is determined by a genetic learning algorithm, in which the second-harmon ic or two-photon fluorescence signal from a reference sample is maximized. The speed of the convergence is improved by use of a Zernike polynomial bas is for the deformable mirror shape. This adaptive optical correction scheme is implemented in an all-reflective system by use of extremely short (10-f s) optical pulses, and it is shown that the scanning area of an f:1 off-axi s parabola can be increased by nine times with this technique. (C) 2000 Opt ical Society of America.