Correlated Imaging Series: Collaborative analysis of multimodal imaging data with the Cytomine web platform
Published: 2022-07-08
On Friday, July 15th at 13:00 CEST, Raphaël Marée, R&D & Senior Research fellow, University of Liège, delivers a lecture on “Collaborative analysis of multimodal imaging data with the Cytomine web platform,” as part of the Correlated Imaging Series, brought to you by COMULIS and Euro-BioImaging.
Abstract:
The Cytomine project (https://www.cytomine.org/ ; Marée et al. Bioinformatics 2016) started in 2010 at University of Liege (https://uliege.cytomine.org) to build a rich web environment for the collaborative analysis of multi-gigapixel imaging data.
This open-source software tool (permissive license) has been designed with the following objectives in mind: provide remote and collaborative principles, rely on data models that allow to easily organize and semantically annotate high-dimensional imaging datasets in a standardized way, efficiently support high-resolution multi-gigapixel images (e.g. whole-slide images in digital pathology), and provide mechanisms to readily proofread and share through the web image quantifications produced by machine/deep learning-based image recognition algorithms. Cytomine collaborative principles allow to break common practices where imaging datasets, quantification results, and associated knowledge are still often stored and analyzed within the restricted circle of a specific laboratory.
In this talk, we will summarize Cytomine’s main features and applications in various bioimage informatics projects, and give an overview of our latest developments for multimodal projects in the context of the COMULIS COST Action. This includes PIMS, our new, extensible, Python-based image management system to support various multidimensional and multispectral microscopy data; and recent novel user interfaces for multimodal annotation. We will also present our extensible software execution architecture that was used in Biaflows for reproducible benchmarking of heterogeneous analysis pipelines (Rubens et al., Cell Patterns, 2020).
Links:
https://people.montefiore.uliege.be/maree/
https://www.linkedin.com/in/rapha%C3%ABl-mar%C3%A9e-739349ba/
About Raphaël Marée:
Raphaël Marée received the PhD degree in computer science (machine learning) in 2005 at the University of Liege, Belgium. While in charge of the GIGA bioinformatics core facility, he initiated the development of the collaborative Cytomine web software in 2010 and he is now Head of Cytomine Research & Development at University of Liège (https://uliege.cytomine.org). His research interests are in web software development, open science, machine/deep learning for image recognition, and their applications to biomedical imaging and other fields that involve big image datasets. He recently co-supervised the development of the BIAFLOWS web platform for reproducible image analysis and benchmarking (https://biaflows.neubias.org/) during the NEUBIAS project (http://neubias.org/). He is co-leading software development work packages for EU projects - COMULIS for correlative multimodal imaging (https://www.comulis.eu/), and BigPicture for digital pathology (https://bigpicture.eu/) - where Cytomine web tool is playing a central role.
Friday, July 15th, at 13:00 CEST
Join via internet: https://us02web.zoom.us/j/760003029 or via phone:
Meeting ID: 760 003 029 Find your local number: https://zoom.us/u/acj97wMY13
