Kick-off H2020 project ExaMode

20 European researchers gathered last week at the Techno-pôle in Sierre, Switserland to kick-off the European H2020 project ExaMode. Coordinated by the Institute of Information Systems of the HES-SO Valais-Wallis, Jeroen van der Laak, Francesco Ciompi and Mart van Rijthoven are working in partnership with the University of Padova (Italy), the hospital of Catania (Italy), SURFsara (Netherlands), Ontotext (Bulgaria) and MicroscopeIT (Poland) on processing of exascale volumes of healthcare data.

The objective of the 4-year project ExaMode is to develop new prototypes for processing large volumes of medical data on exascale computing facilities. These can be used for the analysis of large quantities of histopathological images (100,000x100,000 pixel), including thin-slice images taken with a microscope required for cancer biopsies and other pathologies.

Revolutionising pathological diagnosis with digital technology

More detailed analyses, using artificial intelligence algorithms, could lead to better results, for instance for computer-aided diagnostic assistance with quantitative parameters, often more objective than current processes. Deep learning tools are based on large volumes of training data that has insufficiently been annotated. Generating training data with limited human interaction and using semantics to better understand medical reports and the changes on the images that are described could lead to more powerful algorithms.

Kick-off in Sierre

The project partners gathered for the kick-off meeting at the Techno-Pôle in Sierre, where several research institutes of the HES-SO Valais-Wallis are located. The first day was dedicated to the presentation of the partners and to team-building activities for the consortium, and the second day to the fine-tuning of the project roadmap.