BioImageIT: Integration of image data-management with analysis
Résumé
Open science and FAIR principles are major topics in the field of modern microscopy for biology. This is due to both new data acquisition technologies like super-resolution and light sheet microscopy that generate large datasets but also to the new data analysis methodologies such as deep learning that automated data mining with high accuracy. Nevertheless data are still rarely shared and annotated because this implies a lot of manual and tedious work and software packaging. We present BioImageIT an open source framework that integrates automation of image data
management with data processing. Scientists then only need to import their data once in BioImageIT, which automatically generates and manages the metadata every time an operation is performed on the data. This accelerates the data mining process with no need anymore to deal with IT integration and manual analysis and annotations. BioImageIT then automatically implements FAIR principles. The interest of bioImageIT is thus twice. We
will illustrate this through diverse application workflows, including preprocessing of raw data, complex images reconstructions (i.e Lattice Light Sheet or Multi-Angle TIRF micrscopy), deconvolution/denoising (including DL approaches) and analysis (tracking).
Origine : Fichiers produits par l'(les) auteur(s)