Thread by @CAS_mitolab

Don't have access to a fancy Airyscan or STED?
Do have access to a good ole fashion confocal?
You can still do super resolution imaging!🧵#microscopy

All you need are good microscopy and digital imaging fundamentals. Everything you need to know can be found here: ibiology.org/online-biology… /2

Digital images are made up of spatially defined boxes called 'pixels'. Images are collections of pixels assigned color shades from a lookup table. /3

Here is a very simple image represented by one of four possible shade values. The simple pattern in the image arises from the distribution of pixels in space. /4

Here's a more typical image. Mitochondria stained with TMRM dye. 8-bit depth, allowing 256 possible shade values. Two different lookup tables are presented. /5

A convolution operation systematically uses the neighboring pixel values to modify all the pixel values in an image. This can be used to smooth or sharpen images digitally. /6

Images from a confocal are actually representations of diffraction patterns. Each sub resolution point source of light interferes with its neighbors. /7

If you know the blur pattern of each point source, you can use a fourier transform to correct for the interference. This is known as deconvolution. /8

You'll need to measure the blur pattern (point spread function or PSF) using sub-resolution fluorescent beads. These can be obtained commercially for pretty cheap. The PSF can be calculated from images using MetroloJ plugin in ImageJ. /9

With the PSF in hand, deconvolution can be performed in imageJ using the Deconvolutionlab2 plugin. Here I used the Richardson-Lucy algorithm. The documentation explains all the available algorithms. /10

Here's a close-up of the improvement in signal to noise ratio for a zoomed in portion of the previous images. /12

Here's the final improvement from the original image. /13

Finally, you can push the resolution limit even more by restricting the diameter of the confocal pinhole aperture. This is very nicely explained in this paper👇 Happy imaging!! /14
sciencedirect.com/science/articl…