This enables PSF engineering in space-limited microscopy systems, e.g. that fit into incubators. Uses include extended depth-of-field and snapshot 3D spheroid from 2D, using algorithms by @EliasNehme9. Bonus: Obtaining training images is easy with a high-throughput microscope.

PSF engineering typically requires a 4-f system, but this is not always feasible, e.g. due to size constraints. Nadav Opatovski attached a phase mask directly to the objective lens, see beads defocusing here. Tetrapod mask shown here (wait for the end):

One of these super-resolution microtubule images was captured by a microscope and the other one was generated by a diffusion model. Can you tell which is which? Try more here: clever-microscopy.com/this-m… 1/4

close indeed! also the Schneider logo

2 bonus points if you can guess what this DOE does (clue in the preprint 🙂)

Can you determine your distance from a truck the same way you determine the z position of a fluorescent molecule? Of course you can, all you need is a PSF-engineered telescope! Adventurous Nadav Opatovski built one: doi.org/10.1364/OE.472150

New prepint just out! Live cell single molecule localization microscopy at >3 fps. The information is mostly there - just spread out in space and time :) We use a recurrent neural net (LSTM) + some assumptions to extract it. All details here: biorxiv.org/content/10.1101/…

Field-dependent PSF engineering? Yes! Each 'letter' here is in fact a fluorescent microsphere, appearing differently as a function of its position in the FOV. To do this, Boris Ferdman had to venture beyond the Fourier plane... arxiv.org/abs/2202.11535

Sounds doable. It's easy (and fun) to try it with phase retrieval, e.g. with VIPR (see figure 1b from osapublishing.org/oe/fulltex…).

My favorite auto-caption from a seminar I gave yesterday. The notorious inverse problem of deacon pollution! (or a Captain Planet villain, according to @Lucosaurus)

"Give me enough photons..."

A rotating tilted mirror illuminated by an array of spots. PSF engineering is done only on the detection side. (left - measured, right - recovered) (4/4)

An inflating elastic membrane with scattered fluorescent beads (left - measured, right - recovered) (3/4)

Finally, here is a cool dense tracking example using the Nebulae pair: ~60 telomeres moving around in 3D the nucleus in a live U2OS cell. The two complementary Nebulae channels are shown. (11/11)

But what is the *optimal* PSF-pair for the dense case? We used deep learning to find an optimal phase mask pair. The result is this “Nebulae” PSF pair (thanks @LaineBioImaging for the name!). Information is encoded in the PSF shapes *and* in the relative lateral position (10/11)

Here's one intuitive solution we investigated: Use one PSF to encode z (Tetrapod), and one PSF to encode only x-y, an extended-depth-of-field (EDOF) PSF, devised for this purpose (see gif). Adding an EDOF channel indeed helps. Stealing those Tetrapod photons is worth it. (8/11)