We are actively developing an adaptive-optics system which can be readily integrated into a multiphoton microscope to substantially increase the depth from which images can be obtained.  Such a device is likely to have a wide area of application, as the the biomedical research community is rapidly accepting microscopes capable of obtaining images from deep within living tissue.

Limits to the depth within living specimens from which images may be obtained

In the course of earlier studies comparing the ability of confocal and multiphoton microscopy to obtain images from deep within tissue specimens, we noticed that, although multiphoton microscopy was far superior to confocal microscopy, image degradation inevitably occurs when looking deeply into thick specimens.  This degradation mainly arises from microheterogeneities of refractive index within the specimen giving rise to lensing effects which cause image blurring and distortions.  Multiphoton imaging relies on providing a very high density of photons within the focal volume of the objective lens to produce nonlinear excitation to the fluorophore.  Therefore, any blurring of this focus caused by refractive-index microheterogeneities or lens aberrations will increase the effective focal volume.  This will then lead to reduced photon density, excitation, and resultant intensity of the fluorescence signal.

Adaptive optics

Heterogeneities in the refractive index of the atmosphere impose limits on the resolution that may be obtained by ground-based astronomical telescopes.  Atmospheric non-uniformities distort the wavefront, creating phase errors in the image-forming ray paths.  In recent years, great strides have been made in overcoming this problem for astronomical telescopes through the use of adaptive optics.  Adaptive optics work by measuring the phase front in the aperture plane of the telescope and using this information to modulate the shape of an optical element placed in the light path so that compensatory phase changes can be introduced across the aperture plane.  For astronomical telescopes, there are considerable problems in implementing such a system, not the least of which is that the atmosphere is turbulent and therefore refractive-index heterogeneities will change over time.  Nevertheless, adaptive optics techniques have been successfully applied to astronomical telescopes and have produced spectacular improvements in image quality.

Recently, a few exploratory studies have been published that demonstrate that adaptive optics (AO) can be applied to multiphoton microscopy to increase the penetration depth from which images can be obtained. Multiphoton microscopy may be particularly favorable for the application of adaptive optics techniques because of the nonlinear nature of fluorophore excitation.  Multiphoton events depend on the photon density at the focal volume of the objective lens of the microscope.  For two-photon imaging (most commonly used multiphoton method) the excitation depends on the square of photon density.  Any optical aberrations will reduce the photon density at the focal volume and hence reduce signal.  An adaptive optics system that corrected for sample- and microscope-induced aberrations would restore signal intensity lost due to the blurring of the focal volume.  This suggests that relatively simple optimization schemes could be used to adjust the shape of the phase front and maximize signal intensity.  Multiphoton microscopy has another characteristic that makes it particularly amenable to adaptive optics: the resolution of a multiphoton image is only defined by the geometry of the focal volume of the excitation beam.  Fluorescent photons do not have to be optically imaged, only detected, as they can only originate from the excitation volume.  This means that phase front corrections need only be applied to the excitation path - i.e., the scanning laser beam - and not in the signal path.