Projects

HAF
MDLMA
Autonomous Accelerator


OTHER RESOURCES

IDAF
Blubber


Page tree

Machine Learning is increasingly being exploited on campus in various scientific fields, projects, groups and colleagues. Applications encompass for example autonomous accelerators, segmentation of bio-degradable bone implants. particle physics simulations and track detection, automated processing of electron microscopy images. These pages give a brief overview of AI related activities at DESY, and offer opportunities for groups and projects to present themselves and their research fields.

Note: This is currently a test space without much real content.

Recent Science Highlights

Sample for blogs. See below for the latest three highlights. Add your own scientific highlight!

Photophoretic forces are induced when light causes a net momentum exchange between a particle and a surrounding gas. Such forces have been shown to be a robust means for trapping and guiding particles in air over long distances. Here, we apply the concept of an optical funnel for the delivery of bioparticles to the focus of an x-ray free-electron laser (XFEL) for femtosecond x-ray diffractive imaging. We provide the experimental demonstration of transversely compressing a high-speed beam of aerosolized viruses via photophoretic forces in a low-pressure gas environment. Relative temperature gradients induced on the viruses by the laser are estimated via particle-velocimetry measurements. The results demonstrate the potential for an optical funnel to improve particle-delivery efficiency in XFEL imaging and spectroscopy.

Phys. Rev. Applied 17, 044044

Salah Awel, Sebastian Lavin-Varela, Nils Roth, Daniel A. Horke, Andrei V. Rode, Richard A. Kirian, Jochen Küpper, and Henry N. Chapman

The interaction of intense particle bunches with plasma can give rise to plasma wakes1,2 capable of sustaining gigavolt-per-metre electric fields3,4, which are orders of magnitude higher than provided by state-of-the-art radio-frequency technology5. Plasma wakefields can, therefore, strongly accelerate charged particles and offer the opportunity to reach higher particle energies with smaller and hence more widely available accelerator facilities. However, the luminosity and brilliance demands of high-energy physics and photon science require particle bunches to be accelerated at repetition rates of thousands or even millions per second, which are orders of magnitude higher than demonstrated with plasma-wakefield technology6,7. Here we investigate the upper limit on repetition rates of beam-driven plasma accelerators by measuring the time it takes for the plasma to recover to its initial state after perturbation by a wakefield. The many-nanosecond-level recovery time measured establishes the in-principle attainability of megahertz rates of acceleration in plasmas. The experimental signatures of the perturbation are well described by simulations of a temporally evolving parabolic ion channel, transferring energy from the collapsing wake to the surrounding media. This result establishes that plasma-wakefield modules could be developed as feasible high-repetition-rate energy boosters at current and future particle-physics and photon-science facilities.

Nature 603, 58–62

D’Arcy, R., Chappell, J., Beinortaite, J. et al.

Highly accurate segmentation of large 3D volumes is a demanding task. Challenging applications like the segmentation of synchrotron radiation microtomograms (SRμCT) at high-resolution, which suffer from low contrast, high spatial variability and measurement artifacts, readily exceed the capacities of conventional segmentation methods, including the manual segmentation by human experts. The quantitative characterization of the osseointegration and spatio-temporal biodegradation process of bone implants requires reliable, and very precise segmentation. We investigated the scaling of 2D U-net for high resolution grayscale volumes by three crucial model hyper-parameters (i.e., the model width, depth, and input size). To leverage the 3D information of high-resolution SRμCT, common three axes prediction fusing is extended, investigating the effect of adding more than three axes prediction. In a systematic evaluation we compare the performance of scaling the U-net by intersection over union (IoU) and quantitative measurements of osseointegration and degradation parameters. Overall, we observe that a compound scaling of the U-net and multi-axes prediction fusing with soft voting yields the highest IoU for the class “degradation layer”. Finally, the quantitative analysis showed that the parameters calculated with model segmentation deviated less from the high quality results than those obtained by a semi-automatic segmentation method.

Sci Rep 11, 24237 (2021).

Baltruschat, I.M., Ćwieka, H., Krüger, D. et al.

  • No labels