Moritz Zaiss
@altustro
Followers
503
Following
2K
Media
257
Statuses
877
Heidelberg, Tübingen, Erlangen, MR physicist, science communicator, CEST and Bloch-McConnell fan, Pulseq fan, MR-zero Prof @FAU_Germany
Joined March 2015
MR-zero with phase distribution graphs. #MRI.Git and doc:. Paper: . .
#MRI simulations of images can only be done with spin simulations, but not with EPG? In the new paper Jonathan Endres and Co-Authors show a new MRI simulation called Phase Distribution Graphs which extends EPG for arbitrary timing and image encoding.
0
1
14
RT @RnaudBertrand: I just read this WSJ article on why Europe's tech scene is so much smaller than the US's and China's. I'm afraid that,….
0
1K
0
On the **train** from Nürnberg to ISMRM Hawaii!. Made possible by ISMRMs Minihub initiative in Lille, France.
0
0
2
#FAU MRIpulseq course 2024/25. It was a blast!
Give the students an MRI sequence and they can do one project. Teach the students Pulseq programming and they can do new projects for their whole life. Our #Pulseq course + #MR-zero simu starts again Monday Mar 10, 9:30 (CET). Just join virtually:.
0
0
18
RT @Stikov: This cookbook is long overdue, and it starts with my favorite Buckheit and Donoho quote about academic papers being just advert….
link.springer.com
Magnetic Resonance Materials in Physics, Biology and Medicine - MRI acquisition and reconstruction research has transformed into a computation-driven field. As methods become more sophisticated,...
0
3
0
Lille more conversation, Lille less action. Skip the long-haul flight to Hawaii & join the ISMRM Mini Hub in Lille, France instead! Save time, money, CO₂, and avoid travel stress. Enjoy deep discussions, meaningful exchanges, and a relaxed atmosphere.
0
1
3
This could be the new speed of iteration cycles of science. Tweet about new methods paper. Tweet of a different group reproducing it with new insights, 8h later.
I took a brief look at the Harmonic Loss paper. tl;dr: instead of dot-product with softmax, do euclid dist with normalized 1/d**n. I kinda want this to work. I've dabbled with preferring euclid many times throughout my career (eg triplet loss etc). However. .
2
0
3
RT @schorn_stephan: KI-Fakes sind nicht immer leicht zu erkennen, aber sie enthalten alle irgendwelche Fehler. Hier wird zum Beispiel eine….
0
74
0
Just learned that trainable bilateral filters are most simple transformers.@maier_ak.
aapm.onlinelibrary.wiley.com
Background Computed tomography (CT) is widely used as an imaging tool to visualize three-dimensional structures with expressive bone-soft tissue contrast. However, CT resolution can be severely...
Since we're going back to the origins, let me add that Attention, which @karpathy correctly calls a "brilliant (data-dependent) weighted average operation", was not discovered in machine learning - in fact this dates back to data-dependent "filters" in image processing from the.
1
0
1
The deepdive podcast for the paper is here: a nice gif shiwing the optimization is here:.
github.com
Contribute to MRsources/FLASHzero development by creating an account on GitHub.
0
0
1
Now in #MRM: MR-zero discovers strategy beating quadratic phase cycling for FLASH MRI, leading to an ideally spoiled transient FLASH, removing Epstein-fluctuations. Better PSF, reliable phase, and an actual Look-Locker decay. #weinmüller_cycling #MRI #FAU.
onlinelibrary.wiley.com
Purpose The complex signal decay during the transient FLASH MRI readout can lead to artifacts in magnitude and phase images. We show that target-driven optimization of individual RF flip angles and...
1
0
6
