This page describes the CCMI Deltavision Deconvolution Microscope. This microscope excels with certain types of imaging, especially Z-series, 3D/rotatable reconstructions, time course series, and multiple-color fluorescence microscopy.
|What is deconvolution microscopy?|
|Microscope components and capabilities|
|Information for first-time users|
|The microscope is often busy a week or more in advance, so you may want to keep this in mind when designing your experiments or setting up an appointment to learn how to use the machine.|
Deconvolution is a software-based process by which one can "re-focus" an out of focus image. Deconvolution occurs after image acquisition, and uses nearest-neighbor algorithms to extract information out of blurred regions of an image to clean up these regions - they then appear to be closer to, if not actually in the same plane of focus as the rest of the image.
Taking surface images of 3D objects, as I'm sure you've experienced, can often cause the sort of localised blurring of sections of images described above. A more common problem, however, is the question of what may lie just beyond the current plane of focus you have set for yourself when taking an image of intracellular structures or a cell in the middle of a tissue slice. By merely changing the plane of focus, one can make structures "pop" into view that were previously invisible or at best an informationless blur; this causes a problem for the scientist who wants to know the true layout of structures in their native, three dimensional configuration.
The best solution for this type of problem is to take a Z-series, or a series of images with the same X and Y coordinates but varying up and down the vertical focus (or Z) axis. These images may then be scrolled through later to see all of the pertainent cellular information in sequence, which gives a visual effect akin to becoming really tiny and diving through the cell or tissue slice, only seeing the structures at or close to your current depth as you swim through. Or, an often more exciting (if not always more clarifying) method of presenting the data is to have the computer stack the images to create a single, rotatable image in three dimensions. Either way, you can end up with a lot more information (and concomitantly, a lot less hard drive space) than if you had taken a single two dimensional image.
As you also probably know, deconvolution scopes are not the only microscopes that perform these three dimensional tricks. Many of you, I've found, have heard of or used confocal microscopes in the past, and although deconvolution scopes give a similar product in many respects, the actual mechanism of image acquisition differs greatly, giving each microscope advantages over the other in certain applications. Confocal microscopes use a system of pinholes placed before the collection aperature, which have the net effect of blocking out all light from the sample except for the light originating directly from the current plane of focus. This greatly increases the signal-to-noise ratio in your sample, as you can probably guess, especially when you are looking for single fluorophore molecules or labeled, rare cellular particles in a veritable sea of extraneous (well to you, anyway) light-scattering organelles or other cells. Yet the confocal has its drawbacks - the pinholes must be correctly aligned with the plane of focus to achieve a meaningful image, and for each image in a Z-series, the focus has to be reset.
Not so with a deconvolution microscope. When taking a Z-series, the microscope does not discriminate between light eminating from inside or outside your plane of focus. In fact, you need only to focus the object near the center of the region of interest to you, and then tell the microscope the upper and lower limits of Z-travel, number of images to take in this range, and how far apart (with a resolution capability of 0.1 micrometers in Z) the images should be. You can the let the microscope take the series of images, many of which will be completely out of focus. The software then does its magic (not realy magic, but seems like it!) and reallocates light to the correct pixels, and lo and behold, the images suddenly make sense again, or most of them will make much more sense than they did before, anyway. Still out-of-focus images on the periphery can then be selectively cropped out.
|Raw data image||Same image, post-deconvolution|
A final advantage that deconvolution has over confocal microscopy is that the specimens are illuminated with the white light from a mercury arc lamp that is then passed through a selected optical filter to give the correct excitation wavelength. This also means that to take images of samples stained with multiple fluorophores, the filter wheel must simply be rotated: a much quicker and easier process then changing the actual wavelength of the illumination source. The Institute's confocal microscopes, on the other hand, are illuminated with lasers, and so the availiable range of excitation wavelengths is much smaller. Please see the table below for a full list of the excitation capabilities on this microscope.
Whether the deconvolution scope is suitable for your needs rather depends on your samples. How good is your signal to noise ratio? What have others in the literature used for a similar project? How many image slices do you want to take on the Z-axis? What fluorophores are you using? Will a confocal give you the correct excitation wavelengths? Are you using fluorescent dyes? (Bright-field images work with only limited sucess.) If you are still uncertain, check out Applied Precision's Deltavision page and scroll down to the very bottom for a table comparing the effectiveness of different kinds of microscopy for selected biological applications. Good luck!
|Availiable Microscopy Techniques||Experiment Design Options||Fluorochrome Capabilities||Machine Features||Post-acquisition Software Capabilities|
All first time users MUST be trained in order to use the Deltavision. There is quite a lot of training involved in learning how to use both the microscope itself as well as the software. One of us here in CCMI will have to walk you through at least the first entire session, and possibly even the second, depending how comfortable you have come to feel with the machine. This means you may have to work around our schedule to find a time when the microscope is open and one of us is free to help you, so be prepared.
Especially for your first time, but also for several sessions as you are becoming comfortable with the instrument, you will not be allowed to do live cell work. The Bioptechs live cell chamber is notoriously difficult to setup and work with, and entails quite a bit of frustration for even the experienced user.
On your first visit, David will set up a user account for you on the Silicon Graphics workstations, complete with a individual password. You should not make the password public, of course. If you later decide to instruct someone on how to use the microscope yourself, or bring a labmate along with you to do some work, please be aware that you are responsible for their actions, and any damage that they (hopefully will not) cause to the microscope. Most people, upon finding out how much the microscope costs (go on, ask us when you're down here!) decline to train anyone themselves or let anyone else use their account!
Due to the high demand for scope time on this core instrument, we have had to institute a policy that only one new investigator a week be trained on the instrument. When calling to make an appointment for a training session, please be prepared to be asked to wait a week or more for your appointment - hopefully the line won't be long when you do call. Also in response to heavy use on the scope, and the labor intensive nature of the training session (3-4 hours for a few pictures), we will not train users who just want to learn the scope due to curiosity or if they could simply use a regular fluorescence microscope.
Please review and FOLLOW the rules! A situation wherein you are deemed unworthy to use the Deltavision should never even occur, though, as long as you: