In this webinar, Prof. Rüdiger Rudolf from Mannheim University of Applied Sciences discusses the usage of AI-based segmentation algorithms for cell tracking, particularly in 3D-cell cultures that serve as models for pharmacological drug screening. It is shown how algorithms can be applied to segment challenging objects and datasets using built-in training models.
Prof. Rudolf and the Leica experts detail the collaborative efforts of creating reliable synthetic ground-truth datasets in a semi-automated manner by augmenting the experimental data of nuclei from their own samples. The synthetic dataset is modulated by scattering and diffraction effects, so it better reflects real-world scenarios. The final segmentation yields a 92% detection accuracy for 3D segmentation, which is superior to previously achieved results using conventional segmentation algorithms.
After Prof. Rudolf’s presentation, Leica experts give an overview of the Light Sheet method, as performed on the STELLARIS platform, and show how it can be used to study different samples, such as living and cleared tissues, organoids, and spheroids. Due to the incorporation of a vertical experimental setup, users have more options to configure fixed laser lines, create custom filters, and access samples quickly with multi-position experiment setups.