DPG Phi
Verhandlungen
Verhandlungen
DPG

Berlin 2024 – scientific programme

Parts | Days | Selection | Search | Updates | Downloads | Help

BP: Fachverband Biologische Physik

BP 15: Poster IIb

BP 15.6: Poster

Tuesday, March 19, 2024, 18:00–20:30, Poster F

Insights from live non-linear microscopy imaging: comparative analysis of temperature-induced mitochondrial morphology shifts using standard versus machine-learning method — •Marta Bukumira1, Aleksandra Vitkovac2, Tanja Pajić2, Marina Stanić3, Mihailo Rabasović1, and Nataša V. Todorović41University of Belgrade, Institute of Physics, Belgrade, Serbia — 2University of Belgrade, Faculty of Biology, Belgrade, Serbia — 3University of Belgrade, Institute for Multidisciplinary Research, Belgrade, Serbia — 4University of Belgrade, Institute for Biological Research "Siniša Stanković", Belgrade, Serbia

Using two-photon excited fluorescence (TPEF) modality of our home built nonlinear laser scanning microscope, we investigated, as a proof of principle, mitochondrial morphology adaptations in an eukaryotic cell in vivo, as a response to cooler ambient temperature during growth. The cells were stained with the vital mitochondrial dye Rhodamine 123 in order for mitochondria to exhibit TPEF. We compared cultures grown at two cooler and one control temperature. Acquired images show superior level of clarity and an optimal signal-to-noise ratio, allowing for the morphology differentiations of intricate subcellular structures influenced by subtle temperature variations. Two approaches for extracting parameters from TPEF images were juxtaposed: standard method of particle analysis in ImageJ and nonstandard method in Ilastik, a machine learning-based software. The latter demonstrated greater suitability for this type of analysis, showing increased efficiency in terms of time and reduced susceptibility to errors.

Keywords: TPEF; Imaging; Machine-learning; Mitochondria

100% | Mobile Layout | Deutsche Version | Contact/Imprint/Privacy
DPG-Physik > DPG-Verhandlungen > 2024 > Berlin