Top
SCIENCE. STORIES. IMPACT.

iBiology

Bringing the World's Best Biology to You

Using Software to Control Microscopes

00:00:11.15 In this section, we're going to talk about software
00:00:14.24 controlled microscopes. So, as you're well aware, these
00:00:19.01 microscopes these days are fully motorized. They have digital
00:00:21.29 cameras, they have shutters, you really want to operate all
00:00:26.13 of those components together. Open shutter, take a picture,
00:00:31.05 close the shutter, etc .. And you really need computer control
00:00:35.21 of all the different components. So you need software to run
00:00:39.03 these things. There are a bunch of different software packages
00:00:42.10 out there. One of them is called Micro-Manager. That one is
00:00:49.04 freely available and runs a whole lot of microscope hardware.
00:00:53.01 I'll use that to present software control of microscopes to you.
00:00:58.28 However, all the principles of what's going on here is directly
00:01:03.16 applicable to other software packages, as well. So first of all,
00:01:07.07 we need to load a configuration file. And this is a file that
00:01:13.00 tells the software what hardware is connected to the system.
00:01:18.07 And so I've set that up in advance. And in this case, we have a
00:01:22.11 Zeiss microscope and an old Princeton Instruments camera.
00:01:25.29 And that's about it. And that's all being told here in this
00:01:31.27 configuration file. So I'll load it. Here is the control of the microscope
00:01:40.21 equipment, mainly of the camera. We can set an exposure time
00:01:44.12 and we can snap an image. And so when we snap
00:01:49.03 an image, the software will automatically open the shutter and
00:01:56.00 close it again. And so that is set with this "Auto shutter" check
00:02:01.24 box. If I were to uncheck that and snap an image, we would
00:02:07.05 get nothing. Noise. So since there's no illumination of the
00:02:12.17 sample. But when I open the shutter and then snap an image,
00:02:16.16 it's also fine. But now we're continuously illuminating the
00:02:20.23 sample. So that's not good because we bleach it. So
00:02:25.24 this auto shutter makes sure that your illumination is tightly
00:02:29.24 synchronized with the camera. So I can set an exposure
00:02:34.15 time that is twice as long. And then snap an image. And what you
00:02:41.03 notice is that this image actually looks very, very similar.
00:02:45.03 It didn't get twice as bright. And that is because
00:02:51.10 here in this part, we have set the autostretch checkbox.
00:02:58.27 This part of the screen shows you a -- shows two different
00:03:06.02 graphs all together in one. And it's very useful to have them both
00:03:10.17 together. So one is a histogram that has the grayscale
00:03:15.15 values that the camera can produce, on the x-axis, and then
00:03:20.01 the number of pixels of each of those bins of grayscale
00:03:23.11 values on the y-axis. So from this we can see that the
00:03:28.00 histogram is mainly has relatively low pixel value. So we're
00:03:33.28 very far from saturation of the camera. So seeing
00:03:39.22 directly what your pixel values are is very important, because
00:03:44.03 in principle, you would like to use the full dynamic range that is
00:03:48.14 available on your sensor. So you would want to spread out
00:03:51.20 this histogram as much as possible over the full dynamic range
00:03:56.07 of the sensor. So this shows something else, and that is the camera can
00:04:01.24 produce 4096 different grayscale values. But our poor monitor
00:04:06.28 is not capable of showing all of those. It only has 256
00:04:12.22 grayscale values. Even worse, our eyes cannot really discriminate
00:04:17.29 more than about a hundred or so different grayscale values
00:04:22.01 unless we change the size of our pupils. So, to map
00:04:30.26 the intensities that are available on the camera on the
00:04:35.07 display, we need to have some kind of function. And that
00:04:40.13 function is shown in this same graph. And that is the line going
00:04:45.27 from the dark triangle here to the bright triangle. So, were
00:04:51.15 I to move this line to the left, then my image here gets a
00:04:57.20 lot brighter, because we make pixel values that are lower
00:05:01.24 and lower, whiter and whiter. Were I to bring this left slider
00:05:07.28 here, the dark slider, to the right, then you see that we get more
00:05:13.22 and more black in the image. Now a very useful way
00:05:19.02 to set this up is to use the auto button or to check the
00:05:26.02 autostretch checkbox, so that automatically all the
00:05:29.25 time, the scaling will be applied based on the minimum
00:05:33.16 and maximum pixel values. So you always want to look at
00:05:39.10 this histogram to see what is going on. So for instance, if
00:05:42.29 we now double the exposure time again. So we now
00:05:47.17 go to 300ms, snap an image. Then we see that our
00:05:55.17 histogram moved out to the right. So also our autoscaling
00:06:01.22 moved further to the right. And so in general, the more
00:06:05.25 of your dynamic range, the more of your histogram you're
00:06:09.26 using, the more information you have in the images you
00:06:14.09 acquire. Couple of other important things we want to
00:06:18.28 have a live image. So when I press "Live," we get now a
00:06:24.20 live feed from the camera. And that's for instance
00:06:27.13 useful when you're focusing. So I'm now moving the
00:06:31.14 focus dial on the microscope. And so this way I can
00:06:36.28 focus exactly where I want. Binning. So we talked
00:06:43.28 in the camera lecture about the ability to group groups of
00:06:48.05 4 pixels and read those out in one go. So I can control
00:06:52.15 that here. I can say, okay, give me 2x2 binnings of 4 pixels
00:06:57.10 in one. And when I now snap an image, it is significantly
00:07:02.18 smaller. But you also see in the histogram that it got
00:07:07.04 significantly brighter. Then also what happens is that
00:07:12.02 we're not interested in the whole image, but only in
00:07:15.11 a small region of it. So let's say, this is what we really want
00:07:22.08 to follow. I can use the ROI tool. We only get this small
00:07:29.25 region. So the camera can also read this out much faster.
00:07:34.03 And so that's always helpful to use an ROI whenever it's
00:07:39.21 applicable. Now to go back to the full frame, we use the
00:07:44.15 button next to that ROI and snap an image. All these
00:07:49.27 controls mainly concern the camera, it's very essential.
00:07:54.28 But there are also other parts of the microscope, and
00:07:59.17 in this software, the other parts of the microscope can be
00:08:05.13 configured through these presets. And for instance, I
00:08:10.00 made a preset here for GFP. And what that does is that
00:08:15.05 preset moves the dichroic reflector in place that gives me
00:08:19.23 GFP fluorescence. So likewise, I can move it to Cy3.
00:08:25.12 It moves the reflector in the microscope, I snap an image,
00:08:31.06 and I now have an image of mitochondria, in this case.
00:08:37.16 You will often want to set up experiments where you take
00:08:41.10 these two different channels and do things as
00:08:44.22 timelapses or take z-stacks. That is all hidden under this
00:08:49.01 Multi-D acquisition button. So when I press that, we get
00:08:55.05 a new window. And in this window, we can set up time points,
00:09:00.11 we can set up multiple XY positions, we can set up z-stacks.
00:09:06.00 And we can set up channels. Now these channels, we get from
00:09:11.19 that same group that we just used, the Cy3 and GFP in it.
00:09:18.14 So I'll set the exposure time for this particular channel.
00:09:25.15 Probably something like 400 is good enough. So in this
00:09:34.19 case, we're only using channels. We're taking one image
00:09:38.03 with the Cy3 dichroic and one image with the GFP dichroic.
00:09:42.01 And when I now say Acquire, it first takes the red image,
00:09:50.02 and then the green image, and overlays it for me.
00:09:54.26 And I'll now take z-stacks relative to the current position.
00:10:02.27 And we'll do something like we're going from -3 from where we are
00:10:10.24 to +3 microns, and we'll take steps of 0.5 micron.
00:10:19.20 And then when we acquire now, in this case it has the
00:10:24.05 system has been set up to take the z-stack in one channel
00:10:27.27 first. And it's taking all of those, and then we'll move back,
00:10:32.05 switch to the other channel, and take all images in the
00:10:37.18 second channel. So the focus motor of the microscope
00:10:53.02 makes a half a micron step in between each image. And
00:10:57.22 in the end, we then have a full 3D stack of images. Since
00:11:04.09 this sample is very flat, we basically go through focus here.

This Talk
Speaker: Nico Stuurman
Audience:
  • Researcher
Recorded: April 2012
More Talks in Microscopy Series
All Talks in Microscopy Series

Talk Overview

Modern research microscopes use digital cameras and motorized components and are operated using software. In this talk about using software to control microscopes, Nico Stuurman illustrates some basic principles used in all microscope software packages.

Speaker Bio

Nico Stuurman

Nico Stuurman

Nico Stuurman is a Research Specialist at the University of California, San Francisco, in the lab of Ron Vale. Nico combines his expertise in computer programming and microscopy to advance many projects including the Open Source software, Micro-Manager. Continue Reading

Playlist: Microscopy Series

Reader Interactions

Leave a Reply

Your email address will not be published. Required fields are marked *

iBiology and iBiology Courses are part of the Science Communication Lab (SCL). Our mission remains the same, to connect people to science. However, our focus has shifted to producing and evaluating cinematic films for education and the public, which you can find on the Science Communication Lab website. For more information, please see this blog post!