EXAMODE (EXtreme-scale Analytics via Multimodal Ontology Discovery & Enhancement) is a Big Data project for Healthcare, funded by the European Union’s Horizon 2020 call “Information and Communication Technologies” (ICT-12-2018-2020).
Project website: https://www.examode.eu/
CORDIS website: https://cordis.europa.eu/project/id/825292
Twitter: https://twitter.com/examode
Facebook: https://www.facebook.com/examode.eu/
LinkedIn: https://www.linkedin.com/company/examode/
Contact: Svetla Boytcheva
Project Overview
In today’s landscape of continuously produced volumes of diverse data from distributed sources, Healthcare stands out in terms of the size of the data produced, its heterogeneity, the included knowledge and its commercial value. However, the supervised nature of deep learning models requires large volume of annotated data, which poses a huge challenge and sometimes precludes models to extract knowledge and value.
EXA MODE addresses this challenge by allowing easy and fast, weakly supervised knowledge discovery of exascale heterogeneous data, limiting human interaction. The project involves the development and release of extreme analytic methods and tools that are adopted in decision making by industry and hospitals. It makes use of deep learning in order to build semantic representations of entities and relations in multimodal data.
In EXA MODE, the knowledge discovery is performed via enrichment of text with semantic metadata and the extraction of homogeneous features from heterogeneous images. The results are fused, aligned to medical ontologies, visualized and refined. This knowledge can then be applied to compress, segment and classify medical images and can be exploited in decision support.
Ontotext’s Role
Sirma AI, trading as Ontotext, is part of Sirma Group Holding. In the last decade, Ontotext has developed advanced semantic data normalization components for Life Science and Health Care related projects both in commercial and research settings.
In EXA MODE, Ontotext will lead the work on Multimodal Knowledge Management. We will use text analysis to extract structured data from medical records and annotated images and will fuse them with available structured knowledge. This will provide an innovative approach for semantic data exploration and search.
In addition, Ontotext will contribute to the definition of the semantic model of the domain and will take overall responsibility for the communication about the EXA MODE project to the business community and the general public.
This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 825292.