Motion Detection in Optical Diffraction - Imaging and Dynamic Single Molecule Localization Microscopy

• Subject Area: Mathematics
• Term: since 2022
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 495365311

This proposal covers the fourth year of our sub-project within the SFB "Tomography Across the Scales" handled by the Austrian Science Fund FWF for the whole four-year funding period.In our sub-project, we examine the reconstruction of motion in i) diffraction tomography and ii) cross-registered diffraction-limited and superresolution microscopy. In part i), we have established a new so-called common circle method in diffraction tomography to determine the rigid motion of the object from common circles of hemispheres in the Fourier space. We aim to apply this method to real-world data provided by sub-project 6 (M.\ Ritsch-Marte). To this end, we want to adapt noise-robust numerical algorithms from the well-established common line method from X-ray tomography to our model.In another setting, point markers are moving with the object. Based on the assumption of a smooth object movement, we intend to develop a variational model that incorporates ingredients from sparse matrix factorization. For the numerical reconstruction, we apply fast Fourier transform techniques not restricted to grids. Here appropriate sampling strategies have to be examined and prior knowledge on the object will be incorporated. In part ii), two-color microscopy images are recorded by partners of sub-project 9 (G. Schütz) using single particle tracking in combination with diffraction-limited imaging of T cell receptor microclusters. We investigate to what extent T cell receptor molecules enter, leave, and move within microclusters. We have to improve the particle tracking which will be achieved by optimal transport strategies. We still intend to identify moving microclusters by appropriate features. After successfully retrieving motion parameters on simulated data in our previous work, we will apply and refine this approach for the actual experiment. In particular, our model can be generalized such that the cluster radius may vary between different clusters and over time.

• DFG Programme: Research Grants
• International Connection: Austria
• Partner Organisation: Fonds zur Förderung der wissenschaftlichen Forschung (FWF)
• Cooperation Partners: Professor Dr. Ronny Ramlau; Professor Dr. Otmar Scherzer