Good Seeing


Long the purview of telescopes, the dynamic mirrors and wavefront engineering that enables astronomers to calm the night sky’s twinkle are now founding applications in biological microscopy as well. The techniques, termed adaptive optics, are leading to major improvents in the clarity and depth imaging capabilities of today’s microscopes.

The eyes may or may not be the windows to the soul, but our ocular world plays a central role in how our minds are built. Of all human senses, sight is the most influential to our worldview, the most relied-upon to ascertain the validity of our guesses about reality. Galileo’s observations showed us our place in the cosmos by providing the givens to test the conflicting ideas from Ptolemy and Aristotle against Copernicus and Kepler. Hooke, Leeuwenhoek, and Spinoza seeded the scientific landscape with observations that would provide an alternative to the commonly held belief that disease causes were malarial, that is a result of “bad air,” and preventable by applied fragrance. Even the word “cell,” the fundamental building block of life, was concocted as Robert Hooke viewed the organization of a slice of cork through the lens of a microscope.

Humans are lucky to live under a dense atmosphere, keeping us warm and respiring while it protects us from (most) meteors drawn to our gravity well. The downside is that Earthbound astronomers are like a swimmer watching a birthday party from the bottom of a pool, an assuredly poor choice of viewpoints. The dense, turbulent atmosphere of the Earth confounded observation of especially dim extraterrestrial objects, until a few decades ago when deformable mirrors were introduced to astronomical telescopes to counteract atmospheric aberration. Before adaptive optics, astronomers were limited in the tools available to counteract the atmosphere, and made do by building observation centres at high-altitudes and waiting for “good seeing” conditions to attenuate the blurring of active air. In contrast, a modern adaptive optics telescope can best the resolution of the famous Hubble space telescope, as in the case of the adaptive optics-enabled Magellan II located in Chile.
Astronomers have to look out through the thick soup of the atmosphere, but biological microscopists looking into tissues have even more challenges to contend with. Like trying to peer through a nice cup of milky chai, trying to look at living cells in vivo is a major hurdle to determining the nature of life intact and in action. This has led to the adaptation of the same techniques previously developed to take out the night’s twinkle for use in microscopy, now used to obviate the blur of microscopical imaging at depth.

The problem of imaging through a highly aberrating medium is experienced twice when imaging into tissues: once on the illumination side of the path and again as the signal leaves the sample. Optimising the amount of light that illuminates the desired depth and location and then successfully making it back to a point detector or image sensor determines the clarity and speed with which an image can be formed.

In microscope design there are three characteristics to optimise: temporal resolution (speed), spatial resolution, and depth (signal to noise). Improving one aspect of an instrument invariably leads to a decrease in another quality. Anton van Leuwenhoek made incredible observations using instruments made by carefully melting pulled strands of glass in a flame, and setting the resulting aspherical lenses in pinhole brass frames. The tiny, single lens instruments bear more resemblance to a magnifying glass than to a modern compound microscope. Using one was a matter of holding the entire instrument a few centimetres from the eye and squinting through a tiny aperture at the subject, generally illuminated by sunlight. Game changing innovations that lead the way to improve the overall capability of microscopy beyond zero-sum trade-offs in design optimization are few and far between. It is becoming increasingly clear as the technology matures that adaptive optics is fundamentally enabling for imaging tasks that were not possible with the instruments of a decade ago.

One realm in which adaptive optics is finding ready application is in optical imaging of the living brains of model organisms. Brain imaging was the same inspiration that led Marvin Minsky to invent the confocal microscope in the late 1950s. In confocal microscopy, a pinhole plate rejects the majority of light arising from out of focus areas as the microscope beam is scanned throughout the sample of interest. The pinhole plate ensures that out-of-focus light is rejected, but there is a fundamental limit to how much total optical power can be pumped into tissue before damage occurs. Thanks to the pinhole plate, improving confocal imaging with adaptive optics is straightforward as: any increase in the signal making it to the detector indicates a positive correction for aberrations. Therefore, optimising the dynamic elements of the microscope is a matter of producing the maximum signal.


Beam shaping and point-spread function engineering enable new imaging modalities in microscopy.

The application of adaptive optics to neuroimaging instills a strong sense of the cutting-edge, combining brain science with optical physics, but some areas adaptive optics has made a striking impact may seem are much more domestic. Researchers at Durham Univerity, UK employed adaptive optics confocal microscopy to measure the effects of temperature on the activity of cold-water lipases, enzymes that break down fat and grease. Enzyme names are one of the rare cases of scientific nomenclature being intuitive and informative, the name ‘lipase’ identifies the enzyme’s substrate as lipids, or fats, and the activity as cutting them apart is denoted by the suffix “-ase”. Much like a greasy fingerprint on a pair of sunglasses, the very presence of the substrate induces unwanted blurring amenable to correction with dynamic optical elements.

The state of the art is no longer limited to improving the precision and design of static optics. Dynamic elements allow the microscope and the microscopist to adapt to specific imaging situations, a task for which algorithms and image processing are essential. The computational brains of modern microscopes are integral components of the optical system, as essential as the lenses and mirrors that make up the physical hardware.

In pushing the limits of the types of scientific questions that can be addressed with light, there’s no requirement to generate a two-dimensional image. Rather, the data required to test a given hypothesis may exist as a three-dimensional construct, four-dimensional volume plus time, or even higher dimensionality. Although visualisation of data will continue to be important for science communication, the central role of the image in science may soon take a back seat to generalised, multidimensional data. Paralleling this shift, the next generation of light microscopes will also look radically different than our conventional expectations. The shift has already appeared in commercially available microscopes: the computer is so integral to the light sheet microscope made by German optics giant Zeiss, that the instrument does not have eyepieces. The microscopes we use tomorrow will resemble modern microscopes to the same extent as modern microscopes remind us of Leuwenhoek’s handlenses.


In short order that shiny new confocal system may share the fate of this Leuwenhoek replica. This piece is part of the collection at the Oxford Museum of the History of Science.

Adaptive optics links:

Edit 2014/06/24: Fixed links


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s