Dive into the realm of AI-SPRINT's Personalised Healthcare use case! Explore the complexities of our pioneering approach by navigating through the dedicated sections below:



In AI-SPRINT, the Life Sciences Department of the Barcelona Supercomputing Center (BSC) leads the medical use case on personalised healthcare with its focus on stroke risk assessment and prevention. Over 36 months, BSC implement its COMPSs programming models and machine learning developments in this AI-SPRINT use case. It uses the edge-cloud environment as an effective framework to develop innovative and impactful clinical applications for stroke prevention, with a view to realising the benefits of incorporating wearable technology into healthcare, such as continuous data acquisition and low patient burden.


Use case ecosystem 

This use case combines the efforts of: 

  • Barcelona Supercomputing Center: Data analyst and modelling provider. High performance computing (HPC) provider - MareNostrum. Leading the use case and coordinating all phases of the study from wearable device acquisition to analysis and modelling as a specialist in complex data analytics for personalised medicine and accessibility to high computational power.
  • Foundation Freno al ICTUS: Non-profit organisation with the mission of overcoming the personal, family and social impacts of stroke-related illness. Managing use case study participants and committed to implementing the necessary protocols in compliance with all applicable laws and regulations on personal data protection.
  • Nuubo: Wearable device provider with a team of IT entrepreneurs and doctors specialised in arrhythmia and cardiologists.
  • María Alonso de Leciñana, MD, PhD: Stroke neurologist at “La Paz” University Hospital, in Madrid, serving as a medical advisor in the use case development and implementation. 


Key facts

In the European Union, stroke is the second most common cause of death and a leading cause of adult disability. In 2017, there were 1.12 million incident strokes in Europe, 9.53 million stroke survivors, 0.46 million deaths and 7.06 million disability-adjusted life years (DALYs) lost because of stroke. The number of people living with stroke is forecast to increase by 27% between 2017 and 2047 in the EU (Source: H. A. Wafa, et al, ‘Burden of Stroke in Europe’ in Stroke, vol. 15, No. 8, July 2020, available in open access journal). 
Associated costs are a significant burden in the EU. In 2017, this was estimated at €45 billion, including direct and indirect costs of care provision and productivity loss. These costs are expected to increase dramatically with growing populations and the number of elderly citizens alongside the rise in stroke events and their long-term sequelae. Studies has projected that the absolute burden of stroke is increasing and expected to continue over the next three decades with smaller portions of the population of working age. The imperative is to make greater efforts to prevent stroke and improve healthcare planning and priority setting, with a view to reducing the expected financial and logistic challenges in already strained healthcare systems. (Source: European Journal of Stroke, R. Luengo-Fernandez et al, vol. 5, issue 1, October 2019). 



BSC's pivotal use case within the AI-SPRINT project revolves around personalised healthcare, intricately gathering diverse information types and amalgamating quantitative and qualitative data. Each patient's personalized model discreetly compiles:

  • Lifestyle data from comprehensive questionnaires.
  • Biochemical measurements, encompassing glucose, cholesterol, or haemoglobin when accessible.
  • Digital heart parameters, such as rhythms, atrial fibrillations, and electrocardiograms, extracted from wearable devices.

To manage the distribution and parallelism of resources, BSC advances its high-performance data analytics framework within Edge-to-Cloud platforms. This enhancement enables a seamless integration of AI-SPRINT's technology into a use case specifically tailored for personalized stroke risk assessment and prevention.

AI-SPRINT's primary focus lies in safeguarding sensitive healthcare data and optimizing risk forecasting models. The unique framework enables the incorporation of wearable and mobile devices, leveraging AI to provide valuable insights into patient care. The allocation of workload between cloud and edge is intelligently orchestrated.

Within AI-SPRINT, the Personalized Healthcare use case concentrates on crafting artificial intelligence models for stroke risk assessment using sensor data and lifestyle information. This pilot study enlists the participation of both healthy individuals and those with a history of stroke. The implementation of GDPR-compliant data processing and privacy-preserving techniques, such as federated learning, ensures secure modeling while simulating a scenario mirroring real-world data sharing constraints among different hospitals.



The underlying architecture joins specific design and runtime tools that encompass all the required components and services. The deployable infrastructure consists of Kubernetes clusters and groups of edge devices as 'simulated hospitals', specifically patients' personal mobile phones connected via Bluetooth to individual smartbands.  Secured with SCONE, the deployable infrastructure is used for model training (before and during the pilot study), federated learning, and inference on-demand using OSCAR. At runtime level, COMPSs orchestrates the distribution of the computation, MinIO provides local storage and cloud synchronisation, InfluxDB takes care of monitoring services, IM and EC3 manage container deployment and elasticity, respectively. At design level, SPACE4AI-D selects the resources based on defined performance constraints.




Advances beyond the State of the Art

Advances of our approach to the problem of stroke risk assessment via digital devices covers several aspects. The proposed approach integrates heterogeneous types of information, namely sensors and lifestyle questionnaires, resulting in a wealth of data that can be modelled to obtain an effective subject stratification and, at the same time, insights into the personalised conditions. This will be achieved by exploiting distributed machine learning in a real-time application, leveraging the robustness and resilience of the AI-SPRINT architecture, which will also help avoid common pitfalls related to federated learning, such as the efficiency of model updates communication.


Open source/proprietary

All tools are open source. An algorithm for atrial fibrillation detection embedded in the wearable device (RITHMI, https://rithmi.com/) is proprietary (ARRHYTHMIA ALGORITHM S.L).


Who is this for?

  • Stroke patients and healthy individuals
  • Healthcare professionals treating stroke patients and clinical researchers
  • Big companies, SMEs, start-ups working in the stroke and digital medicine sectors
  • Stroke associations/organisations
  • National/international entities with a prominent role in stroke research and stroke-focused societal activities
  • Organisers of stroke prevention and awareness-raising campaigns, events, educational activities
  • Regulatory bodies producing guidelines and accreditations related to stroke and digital medicine
  • National/international stroke-related projects carried out by administrations, organisations, public or private entities


Business Impact

The final model will provide personalised notifications, alerts, and recommendations for stroke prevention, which is a reduced number of strokes, and a faster detection of risks of strokes.
This project, based on Artificial Intelligence in the healthcare industry, represents a unique opportunity for implementing effective healthcare applications working in real-time monitoring patients’ vital signs. Using Artificial Intelligence to help reinvigorate modern healthcare, such as:

  • Improved patient monitoring.
  • Enhanced patient care.
  • Reduced human error.
  • Better managed patient flow.



  • Continuous and non-invasive monitoring over long periods of time
  • Identification of early indicators of health issues
  • Management of at-risk and vulnerable people
  • Personalisation of technological solutions matching individual needs
  • Patient empowerment and information about their own health
  • Patient engagement in healthier lifestyle and behaviour change
  • Reduced strain on the healthcare system
  • Accelerated digitisation of healthcare
  • Personalised care treatment plans
  • Innovation and growth of the wearable tech industry
  • Improve confidence in digital health through privacy and security



Potential societal or environmental impacts

The BSC personalised healthcare use case will significantly impact on well-being. It will pave the way for an effective framework based on Artificial Intelligence models helping to prevent stroke risks coupled with a lifestyle data. This is expected to benefit people aged between 40 and 80, improving and extending human lives. AI-SPRINT will therefore also contribute to the United Nation’s Sustainable Development Goals (SDG 3) on “Ensuring healthy lives and promoting well-being for all at all ages”, including increasing life expectancy and helping to save lives. 


Connect with AI-Sprint

Twitter  |  LinkedIn