41 HIGHLIGHTED PUBLICATION CELL IMAGING & FLOW CYTOMETRY Thanks to FEDER grants, the platform acquired a new STELLARIS 8 (Leica) confocal microscope, equipped with a white laser that allows for excitation wavelengths to be selected within a wide range from 440 nm up to 790 nm. The microscope includes Hybrid S and X detectors for better sensitivity and new modes of signal detection that are more accurate. Additionally, the TauSense technology available on the microscope allows the generation of images using lifetime-based in-formation. This technology requires smaller data size and computational load compared to fluo-rescence lifetime imaging (FLIM). Thanks to FNRS grants and in collaboration with GIGA-Immunohistology, the platform acquired a high-performance AXIOSCAN 7 (Zeiss) slide scanner, for fluorescence (7 colors), brightfield and polarization. The scanner can digitize up to 100 slides in a single run in a fully automated 24/7 operation. New development in confocal spectral microscopy . Classic channel-like acquisition can distinguish only up to 4 fluorochromes. Spectral imaging coupled with linear unmixing has been developed to overcome this limitation by detecting more fluorescent probes and resolving overlapping fluorescent spectra. Machine Learning: Pixel classification with QuPath. Cell detection (segmentation) in microscopy can be challenging. Training a pixel classifier provides more information than a simple threshold and allows for a more sophisticated determination of the output. This method can be applied to cases where a threshold is not accurate enough. Deep Learning: example segmentation. Cellpose is an AI algorithm that can segment phase or holotomographic microscopy data without labeling. It enables quick analysis of live-cell images to study cellular processes, morphology, etc. Cellpose can handle different imaging modalities and has an intuitive interface for ease of use. This breakthrough in AI-based microscopy analysis has the potential to revolutionize cellular process research.
RkJQdWJsaXNoZXIy MTk1ODY=