Bridging the gap between fundamental computer vision research and drug discovery

EU Horizon 2020 research and innovation project award for IMED Discovery Sciences team

Yinhai Wang and Claus Bendtsen, two colleagues from IMED Discovery Sciences, recently secured an award from the European Commission’s Horizon 2020 project. As a result, 12 academic researchers from 6 mainly European Institutes will be seconded to AstraZeneca for a total of 33 months over a period of four years, starting in March 2018.

The programme, entitled Nonlocal Methods for Arbitrary Data Sources (NoMADS), is part of the Marie Curie Research and Innovation Staff Exchange (RISE) programme. RISE encourages bright academic staff to work, interact and share knowledge with industry. AstraZeneca is one of 24 academic and industrial partners under NoMADS.

Methodologies such as deep learning, graph partitioning and spectral clustering have gained increasing attention within the field of Computer Vision in recent years. The combined use of these methods is expected to herald the next wave of breakthroughs in artificial intelligence and industrial automation. Nonlocal methods help solve ‘early’ computer vision problems, including image de-noising, enhancement and segmentation. Research in this area has picked up speed with the development of better mathematical models that fit in a wider field, regardless of modalities and image dimensions.

We are always on the lookout for state-of-the-art imaging technologies that enable our scientists to see crystal-clear images, analyse images with better accuracy, and reveal hidden biology and modes of action from complex multi-dimensional, multi-modality imaging data. I congratulate the team on this successful award.

Mike Snowden VP, Discovery Sciences, IMED Biotech Unit

Imaging technology plays a key role in progressing our drug discovery pipeline. This includes microscopic imaging technologies (multiplexing, transmitted light, phase contrast, and tissue imaging); traditional CT, MRI, and PET imaging; and cutting-edge mass spectrometry and super resolution imaging. Improvements in fundamental computer vision methodology will significantly impact these imaging areas. And by further combining this with the use of deep learning, we can enhance our ability to deliver our pipeline.

Yinhai Wang Discovery Sciences, IMED Biotech Unit

This EU H2020 project, one of the first of its kind, engages fundamental computer vision research with its direct application domain – biological imaging. Rather than merely producing high quality computer vision publications, this project will directly utilise the mathematical innovation for drug discovery and will contribute to reducing our cycle times. I encourage my department and other colleagues to engage more widely with enabling projects such as this and to strengthen our culture of not only following the science but also innovating in-house.

Claus Bendtsen Discovery Sciences, IMED Biotech Unit

About Horizon 2020

Horizon 2020 is the biggest EU Research and Innovation programme ever with nearly €80 billion of funding available over 7 years (2014 to 2020), in addition to the private investment that this money will attract. It promises more breakthroughs, discoveries and world-firsts by taking great ideas from the lab to the market. Learn more

  • Join us

    AstraZeneca is one of the world’s most exciting bio-pharmaceutical companies. Be inspired by the possibilities of science to make a difference and discover what you can do.

  • Partner with us

    We know that innovation doesn’t happen in isolation, so we partner with others to access the best science and contribute complementary technologies, know-how and molecules.


  • Science