BSC to help deliver high performance for groundbreaking healthcare applications in DeepHealth

28 May 2019

Combining massive computing power with big data analytics is radically transforming healthcare, helping accelerate scientific discovery and innovation. In the DeepHealth project, funded by the European Commission’s Horizon 2020 programme, powerful computing resources will be harnessed for biomedical applications. By applying deep learning techniques to large, complex biomedical datasets, the project will support more efficient diagnosis, monitoring and treatment.

DeepHealth will develop a flexible, scalable framework for the high-performance, big data environment, based on two new libraries: the European Distributed Deep Learning Library (EDDLL) and the European Computer Vision Library (ECVL). Fourteen use cases will be used to train models with data from different medical areas, including migraine, dementia and depression. The resulting trained models and libraries will be integrated in seven existing biomedical software platforms, in both the commercial and research arenas.

Barcelona Supercomputing Center (BSC) is playing a key role in this innovative project by enabling the deep learning and image processing libraries to efficiently use high-performance computing (HPC) infrastructure resources. This work will be based on COMPSs, the BSC superscalar framework that facilitates the development and execution of applications for distributed infrastructures such as clusters, grids and clouds.

As part of its contribution, BSC will support the execution of DeepHealth’s libraries on heterogeneous computing architectures including graphics processing units (GPUs) and reconfigurable elements (FPGAs), allowing DeepHealth’s applications to exploit maximum computing power.

‘Our contribution to the DeepHealth project builds on BSC’s proven expertise in fully exploiting high-performance computing resources,’ says Eduardo Quiñones, senior researcher in the BSC Computer Sciences Department and BSC principal investigator in the DeepHealth project. ‘With DeepHealth, we will help deliver the high-performance big data analytics which will power the biomedical discoveries of tomorrow.’

Led by Everis, launched on 1 January 2019 and lasting for three years, DeepHealth brings together a multidisciplinary team of 21 partners from nine European countries, including research institutes, health bodies and companies large and small.

Further information:

About DeepHealth

Funded by the European Commission’s Horizon 2020 research and innovation programme, DeepHealth is a three-year Innovation Action with a budget of over €14 million. The project will put high-performance computing power to the service of biomedical applications; and apply deep learning (DL) techniques on large and complex biomedical datasets to support new and more efficient ways of diagnosis, monitoring and treatment of diseases.

The project brings together a multidisciplinary team of 21 partners from nine European countries, including research institutes, health bodies and companies.