Tautvydas Kiršys

Societal challenges

Audiences are consuming more and more video, demanding increasingly higher quality, using a variety of devices including TVs, smartphones, tablets and computers. 

This is why video compression standards are needed, which allow compressed content to be distributed and then decoded by anyone – ready to be displayed on the device of choice.

The video coding team within BBC R&D focuses on multiple aspects of video technology, with the general goal of supporting the delivery of high-quality content to all BBC audiences. In addition to performing core fundamental research on video compression standards, the video coding team is researching new, advanced ways of performing compression based on machine learning, artificial intelligence and content analytics, while also applying our findings to enable new content experiences.

Technical challenges

The use of large-scale processing resources have the capabilities to transform how content providers obtain, produce and deliver content in challenging scenarios. A move away from expensive bespoke broadcast specific facilities and hardware to more commoditised scalable-cloud based resources will enable providers to more efficiently manage its content compared to what has traditionally been achievable.

How EOSC can help and add value

This use case is part of the EOSC Digital Innovation Hub, which supports its implementation and facilitates access to the EOSC services to the private sector. BBC is advised and supported in the use of scalable GPU compute platforms provided by European e-infrastructure providers. 

Societal challenges

Space Weather concerns the phenomena that arise from the changing physical conditions of the Sun and its effects at Earth. It may seem an abstract topic, but events such as geomagnetic storms have actual practical and economic consequences at Earth, especially as our economies are increasingly dependent on technology and satellite communications and services. The possible consequences and damages due to Space Weather can be mitigated with appropriate protocols that rely on accurate forecasts.

Technical challenges

The DRACO project aims to establish a planetary network of observatories capable of generating high-resolution cosmic ray Space Weather data with an unprecedented level of detail. An advanced cloud infrastructure is essential to manage the distribution and ensure the availability of this Space Weather data. 

How EOSC can help and add value

The DS-DRACO pilot, developed by Hidronav as an EOSC DIH business pilot, was the first step to develop such data infrastructure. The underlying DS-DRACO pilot infrastructure is now storing and providing access to high-resolution data from and for early adopters. 

The research-facing web portal is also online, as a hub of interoperable data for Space Weather modelers and forecasters. Future Space Weather products derived from raw data will be distributed through this platform.

Societal challenges 

Web services owners struggle daily to protect their websites from bot traffic and their users from fraudulent digital ads or cryptocurrency web mining. The problem is not going away: recent years show an increasing number of bot attacks in the global network. For example, considering just online ads, the Association of National Advertisers and White Ops estimates that in 2016, bots were responsible for seven billion dollars of wasted resources.

Technical challenges

The solution to this problem is to “know thy website” via analytics and in-depth statistics that provide insight on the website traffic, and turn those insights into defenses from botnet attacks.

As part of their collaboration with the EOSC-hub project, Koma Nord and Idego designed and developed Guardomic – a tool suite to protect online services from botnets attacks. Guardomic also allows to analyze and block unwanted traffic without decreasing performance for the reader. 

The team needed a flexible server and storage platform that they could remotely access, configure and manage during the development of Guardomic. As an EOSC-hub pilot, they were able to use the cloud compute resources provided by PSNC, namely their OpenStack cloud platform, as a development environment and a production platform for the clients testing the prototype.

How EOSC can help and add value

Working within EOSC-hub has allowed Koma Nord and Idego to develop and implement Guardomic quicker and more efficiently. As part of the EOSC Digital Innovation Hub, they are also now part of large European consortia of science institutes, companies and other organizations, which brings added opportunities for business and extended collaborations. Koma Nord will also take advantage of the EOSC Marketplace as a platform to promote Guardomic as a solution to mitigate botnet attacks.

Societal challenges

The Spanish Foundation for the Study and Treatment of Gaucher Disease and other Lysosomal Diseases (FEETEG) supports the scientific research of Gaucher disease and its treatment methods. The Foundation is interested in predicting the probability of development of diseases such as neoplasms or Parkinson’s disease in patients of Gaucher disease (correlations between diseases). For this purpose, Kampal Data Solutions was contacted by FEETEG to develop an advanced analytical model based on Artificial Intelligence with the information available in the Gaucher Spanish Disease Registry. 

Technical challenges

Due to the fact that Gaucher disease is a rare disease with few national registries, the computational power of a local computer for the study of correlations with other diseases was enough to analyse the data collected. The challenge is to generate a new model able to predict if a person has the probability of developing Gaucher disease. In this case, the AI model must include not only data from current Gaucher disease patients but also data from healthy patients. Opening our sample universe also to healthy patients exponentially increases the sample size and potentially the model’s complexity. This implies the need of advanced computational resources such as the cloud platform provided by EOSC.

How EOSC can help and add value

In the context of the EOSC-hub project, Kampal Data Solutions is benefitting from storing the healthy and ill patients’ registries to a database on EOSC infrastructure. EOSC also supports the pilot as it statistically analyses the data and develops a classifying model based on machine learning techniques. It is also optimising a machine learning algorithm for a cloud based environment and validating the model performance and producing plots /charts of diverse KPIs.

Societal challenges

We want citizens to measure their environment by using smartphones. Most of the citizen science initiatives are focused on one topic: it can be air quality, biodiversity. All in different apps and communities, with different logins and data storage. These apps and communities are mostly project based and therefore time limited. We offer the mobis service: one starting point to collect both environmental and biodiversity data in many forms.

We had the idea of taking sensor measurements and combining them by running (run4science.org). During our trail run, we collect air quality, water quality (color, spectral properties) and take pictures of lycens. For example, we ran around our town and all the data ended up in the MOBIS framework.

Technical challenges

Integrating different observations is challenging. Some are coming from data entry, some from bluetooth devices, some from direct sensors (like smartphone camera/gps). We also want to guarantee that data is processed along GDPR guidelines and licensing. A solution is to partner with other EOSC services (like Authenix for safe and privacy friendly Single Sign On).

How EOSC can help and add value

By offering our MOBIS observation integration service (sensor) app developers do not have to reinvent the wheel. Integration, GDPR compliance, powerful storage and off/online processing are already created. EOSC makes it FAIR!

We will also work together (technically) with other EOSC services, like Authenix and Plant*Net API.