Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe
BP: Fachverband Biologische Physik
BP 12: Poster 2
BP 12.21: Poster
Dienstag, 6. September 2022, 17:30–19:30, P4
New directions in traction force microscopy — •Johannes W. Blumberg1,2, Timothy J. Herbst3, Ullrich Koethe4, and Ulrich Schwarz1,2 — 1Institute for Theoretical Physics, Heidelberg University, Germany — 2BioQuant, Heidelberg University, Germany — 3German Cancer Research Center (DKFZ), Heidelberg, Germany — 4Visual Learning Lab, IWR, Heidelberg University, Germany
In traction force microscopy (TFM), the mechanical forces of cells adhering to an elastic substrate are estimated from the substrate displacements as measured by the movement of embedded marker beads. While it is straightforward to calculate the deformation field resulting from a given traction pattern (direct problem), it is challenging to estimate the traction pattern from the deformation field (ill-posed inverse problem). Usually, an estimate is obtained by minimizing the mean squared distance between experimentally observed and predicted displacements (inverse TFM). Here we explore two alternative approaches in TFM. First, we compare inverse TFM to the direct method, in which the stress tensor is calculated directly from the displacement data, thus avoiding the use of a loss function. Second, we explore the potential of machine learning and convolutional neuronal networks. By applying recently developed conditional invertible neuronal networks (cINN), we can address questions regarding the stability and uniqueness of the obtained traction field estimates.