Wiebke Jahr
LightNews LightNews
banner
wiebkejahr.bsky.social
Wiebke Jahr
@wiebkejahr.bsky.social
180 followers 300 following 23 posts
Physicist, microscopist, tinkerer, builder, programmer 🔬🌈
Alumnus: @HFSP longterm fellow, @ISTAustria postdoc, @huiskenlab @mpicbg PhD.
#LINO19
Now making 3D printing #BetterWithOptics
Posts Replies Media Videos
wiebkejahr.bsky.social
Us two days ago. Dropped mine off while on a business trip in Munich.
Seems that all votes picked up by the last collection on Saturday will still count.
Doesn't matter which party, it's only important that you cast your democratic vote!
February 20, 2025 at 5:09 PM
wiebkejahr.bsky.social
The control suite supports different optical geometries:
- single pass
- double pass, where the beam bounces off the SLM twice to imprint two phase patterns (on orthogonal polarisations)
- split, where each half controls independent beams, to be recombined after the SLM (or not)
Technical diagram labelled as "single pass". The diagram shows a beampath reflected off an SLM once. The helical vortex phase pattern is shown beneath the SLM using a false color image. The incoming beam is labelled as "Gauss" and the two intensity crosssections (xy and xz) are shown. The outgoing beam is labelled as "donut / 2D STED". Also here intensity crossections are shown. Technical diagram labelled as "double pass". The diagram shows a beampath reflected off an SLM, then passing through a lambda/4 plate onto a mirror, again through the lambda/4 plate and back to the SLM, where it is reflected a second time. Beneath the SLM, a helical vortex phase pattern and a tophead phase pattern are shown as false color images. The incoming beam is labelled as "Gauss" and the two intensity crosssections (xy and xz) are shown. The outgoing beam is labelled as "donut / 2D &3D STED". Both intensity cross sections for 2D and 3D STED are shown. Technical diagram labelled as "split geometry". The diagram shows two complicated beampaths, one in red, one in blue. Both are reflected off the same SLM, separated by a triangular knife edge mirror and recombined using a polarizing beamsplitter. Like in the other two diagrams, the helical vortex phase pattern and a tophead phase pattern are shown as false color images beneath the SLM. The incoming beams are labelled as "Gauss" and the two intensity crosssections (xy and xz) are shown. The outgoing beams labelled as "donut / 2D STED" and "donut / 3D STED", respectively. Both intensity cross sections for 2D and 3D STED are shown.
January 15, 2025 at 3:31 PM
wiebkejahr.bsky.social
To ease your (and my) lives, I calibrated all calculations to accept microscope objective specs, I support swappable objectives & implemented a few typical Nikon options already. Configurations can be permanently stored & retrieved; I am using the human-readable .json format.
Screenshot of a GUI showing a dropdown, labelled for different typical microscope objectives (100x Oil, 100x Silicon, 60x Water, 20x Air).
January 15, 2025 at 3:31 PM
wiebkejahr.bsky.social
Simulated mode is perfect for testing with no hardware attached!
Me & intern Hope McGovern used it extensively to train a neural net to recognize aberrations, with the goal to auto-align the real microscope.
I presented the results at FOM2021:
www.focusonmicroscopy.org/past/2021/PD...
Presentation slide showing a workflow for adaptive optics for microscopy. On the left side is the "in silico training data generation", showing the phase patterns and simulated intensity distribution. In the middle is a PC screen pictogram with a stylized neural net, labelled as "train neural net". On the right side is the "in-situ aberration correction", showing again the phase patterns and intensity distributions, but this time as they would be displayed in the microscope.
January 15, 2025 at 3:31 PM
wiebkejahr.bsky.social
Your optics aren't perfect?
The predominant aberration-modes, eg astigmatism (left) or coma (right) are corrected simply by dialing in the respective corrections.
The SLM surface curvature is compensated by uploading the flat-field correction image (provided by the supplier).
False color image, split in the middle. The left half shows a diagonal cross with lines bent outwards in the quadrants. Right half show an image with two half-circles in the center (red/blue), surrounded by tear-drop shaped lines. Red/Yellow/Blue false color image with a diagonal oval shape.
January 15, 2025 at 3:31 PM
wiebkejahr.bsky.social
The GUI gives access to all relevant parameters and displays a preview of the pattern displayed on the SLM.
First screenshot shows “vanilla” setup with 2D STED on left SLM half, 3D STED on right, no corrections. Second screenshot is with some typical corrections applied.
GUI with many numeric controls. In the middle, the phase pattern on the SLM is shown. The pattern is split in the middle. Both halves show a diagonal grid. The left half has a point-singularity in the center, the right half a circle. The same GUI. This time, some of the numerical controls are non-zero. The diagonal grid on the left half is bent. The diagonal grid on the right half is heavily distorted.
January 15, 2025 at 3:31 PM
wiebkejahr.bsky.social
Need more options?
I implemented a code-input dialogue for prototyping and testing, eg with the concentric circles with increasing amplitudes.
Not confident enough to code (yet)?
Just upload your vortex pattern as an image instead & dial in all other corrections via the GUI.
Four images: A false color image showing concentric rings of increasing intensity A heatmap showing more concentric rings, this time with alternating minima and maxima Two identical heatmaps showing a heavily aberrated PSF
January 15, 2025 at 3:31 PM
wiebkejahr.bsky.social
The most commonly used vortex patterns can be selected from a drop-down menu.
Images always show the pattern on the SLM and a simulated intensity distribution:
- Gauss ("plain" confocal)
- 2D STED (donut)
- 3D STED (bottle beam)
- Segments (easy STED donut)
- Bivortex (2D + 3D)
False color images aligned in a grid. The rows are labelled - Gauss - 2D STED - 3D STED - Segments (Easy STED) - Bivortex Columns are labelled as - Vortex pattern on SLM - Intensity distribution in focus -- xy section -- xz section -- yz section
January 15, 2025 at 3:31 PM
wiebkejahr.bsky.social
🔬peeps!
I am delighted to finally publish my code for controlling a spatial light modulator (SLM) to perform adaptive optics & to sculpt the vortex beams for STED & MINFLUX microscopy - complete w graphical user interface for easy use!
github.com/wiebkejahr/s...
Short user manual🧵
Image showing a GUI with many numeric inputs on the left side and color coded intensity maps, illustrating a workflow, on the right side.
January 15, 2025 at 3:31 PM
wiebkejahr.bsky.social
Weekend escape from never-ending Viennese winter-fog to sunny Rax.
Super happy to see the first bit of sugar-coating on the mountains ❄️
But the best treat was the frozen waterfall. Simply majestic!
Schneeberg, the highest mountain in lower Austria, beneath a brilliant-blue sky, is seen thru a curtain of icicles. A section of vertical rock-steps, forming stairs. It's covered by moss and a curtain of long icicles. In summer, there would be a waterfall, but it's frozen in place. Wide path covered with a very light layer of snow.
November 17, 2024 at 8:03 PM