High Speed Imaging with the Andor Zyla

We’ve been having some more fun with our Andor Zyla. While it runs at 100 fps when you image the full field of view, if you work with smaller regions of interest (ROIs), it goes quite a bit faster. The fastest we’ve had it running is 2000 fps for a 1024 x 64 ROI, and it runs at 420 fps for a 528 x 512 ROI. Here’s a movie of a swimming Tetrahymena recorded at 420 fps. It’s been slowed down by 10-fold for display here.

Swimming tetrahymena, recorded at 420 fps with an Andor Zyla camera using a 100x / 1.4 NA lens and DIC optics.

Of course, being a microscopy geek, the second thing I decided to do with the high speed imaging capabilities of this camera was to probe the performance of the microscope itself. Below is a montage of images acquired at 2.5 msec intervals during a 5 micron step of the piezo Z-stage.

Images of a tissue section acquired with a 20x / 0.75 NA lens. Frames are acquired every 2.5 msec. At frame 4, a 5 micron step is made with the piezo Z-stage. It takes 5 - 10 msec for the piezo to settle to its final position.

Images of a tissue section acquired with a 20x / 0.75 NA lens. Frames are acquired every 2.5 msec. Between frames 3 and 4, a 5 micron step is made with the piezo Z-stage. It takes 5 – 10 msec for the piezo to settle to its final position.

With this high speed acquisition, we can clearly see that the piezo does not move instantaneously. However, it’s still quite fast. We then applied this same approach to look at the speed of the Nikon Perfect Focus System (PFS), by taking the same 5 micron piezo Z-step with the PFS on. Under these conditions, the stage moves the sample, and the PFS then moves the objective to compensate for this change in focus. A movie of this process, slowed down 10-fold, is shown below.

A 5 micron Z step is taken with the piezo Z-stage and the PFS moves the objective to compensate.

As can be seen here, the piezo movement is very rapid and the PFS response to this movement is considerably slower. Presumably, the slow response is not due to the PFS itself, but rather to the speed with which it can move the objective. Because the objective turret is heavy, moving it is relatively slow. To more accurately measure the PFS response speed, I quantified the focus by measuring the standard deviation of the image within a small ROI.

PFSresponse_graph

Focus response from the movie above. The PFS takes several hundred milliseconds to return to the focal plane after the 5 micron shift.

Among other things, this explains why we haven’t had great luck using the PFS to keep samples in focus when we do high speed image stitching. The stage is simply moving too fast for the PFS to keep up when we image a new field of view every 100 milliseconds. One potential solution to this would be to integrate a hardware autofocus system with a piezo Z-stage, which is something I’d like to try in the future.