H2020 grant for Computational Pathology

29 November 2018

Jeroen van der Laak and Francesco Ciompi have been awarded an ICT12 H2020 grant of € 4.3M with € 675K for Radboudumc as part of a consortium of partners.

Exa-scale volumes of diverse data from distributed sources are continuously produced. Healthcare data stand out because of size, heterogeneity, the information present in the data and its potential commercial value. Healthcare data produced in 2013 were 153 exabytes, while the data produced in 2020 will be over 2’000 exabytes, posing a challenge for clinicians who need to examine these volumes. Healthcare data are highly heterogeneous, since they include various types of images and signals, time series and medical reports. Text is a unique resource because medical reports carefully describe the diagnosis associated to multimodal and multimedia data. Moreover, text is also used in scientific papers and blog posts to describe specific images, thus making healthcare related data potentially the biggest annotated data collection worldwide.
EXA MODE’s overall aim is to allow easy and fast, weakly supervised knowledge discovery of exa-scale heterogeneous data in the field of digital pathology. Knowledge extracted from data will be used to build computer algorithms based on artificial intelligence, to assist pathologists in routine diagnostics.

EXA MODE is a consortium consisting of academic research groups and companies from several European countries: Netherlands, Switzerland, Italy, Bulgaria, Poland. Together with Radboudumc, the national supercomputing center SURFSara is part of this consortium. ICT12 is an EU program in the Horizon 2020 framework for financing projects on “Big Data technologies and extreme-scale analytics”.

Jeroen van der Laak is member of theme Women's cancers.
Francesco Ciompi is member of theme Tumors of the digestive tract.