Confocal microscopes use a scanning laser epi-illumination to excite fluo-rochromes within the specimen and detect the resulting fluorescence by means of a photomultiplier tube (PMT). The performance of the microscope is ultimately limited by the quality of its conventional optical components, and the same considerations for optimization exist as for conventional fluorescence. Confocal microscopy requires objective lenses of high NA. The range of possible excitation wavelengths depends on the sophistication (and hence cost) of the laser light source. As for conventional fluorescence imaging, filtering both the excitation and emission spectra relies on dichroic beam-splitting mirrors.
The advantage of confocal microscopy is in the ability to focus on a defined optical plane at a given depth within the tissue. This is achieved by collecting light through a small pinhole in front of the PMT. Reflected light from outside the focal plane diverges and falls outside the pinhole aperture. The smaller the pinhole, the narrower the thickness of the optical section and the greater the spatial resolution. The optical section gives a very clean image without interference from scattered light outside the plane of focus. However, narrow optical sectioning also results in an image that is much dimmer than a conventional fluorescence image. Full fluorescence intensity is restored when thin optical sections are taken in series through a specimen and then digitally recombined to produce a projected image through the entire depth of the tissue. Since each element of the resulting projection is in sharp focus, there will be less background than with a conventional fluorescence image.
PMTs can also be used to collect bright-field pictures by collecting transmitted light gathered through the condenser lens. Such images lack any color, contrast information and their resolution depends on the optical quality and correct focusing of the condenser.
The aim of confocal microscopy is the production of an optimized black-and-white digital image. Subsequent image manipulation can enhance the image by addition of pseudocolors, but the amount of information must be maximized at the point of image acquisition. The quality of the PMT determines the resolution (number of pixels) and number of gray shades in this final image (for example, an 8-bit image has 256 gray levels). However, regardless of the type of PMT, the aim is always to utilize the available gray levels fully. The points of interest in the image must be spread over the maximum tonal range. In some cases, this may involve setting a nonlinear collection ramp such that background is either suppressed or the intensity of faint objects within the image enhanced. Performing nonlinear filtering after the image is collected will only result in a loss or distortion of the original tonal information. Optimizing the image always involves setting an arbitrary "black-level" at an intensity of 0, and ensuring that the brightest elements of the image fall just below or at the maximum available pixel brightness (saturation point) by setting a gain control. Too much gain produces a saturated image that contains very little tonal information and appears very "flat." Where the fluorescence emission is low, increasing gain also leads to increased noise levels in the PMT. Noise can be compensated for by collecting a final image that is an average of successive scans of the same view. The number of scans to be averaged depends on the gain of the PMT and can be set empirically by determining how many averaged passes are required before the image stabilizes. These principles also apply to other digital image capture devices.
Was this article helpful?