BSC develops AI model to predict stroke risk using mobile devices

08 November 2024

The Innostroke project aims to transform the prevention and monitoring of stroke, one of the leading causes of death and disability worldwide, through artificial intelligence

The Innostroke technology has been developed in an interdisciplinary way by the BSC's Computer Sciences and Life Science departments, with the participation of diverse profiles such as biotechnologists and computer engineers.

The technology is based on the use of electrocardiogram (ECG) data collected from mobile devices such as smartwatches and incorporates lifestyle information through a mobile application.

In collaboration with the Hospital de Sant Pau, the project seeks to incorporate a third dimension of stroke risk, genetic and molecular biomarkers, with the aim of offering personalised and proactive medical care that improves patients' quality of life.

Within the scope of digital health and personalized medicine, the Innostroke’s project technology developed by BSC researchers focuses on using digital tools and artificial intelligence to deliver more efficient, precise, and personalized healthcare. Three main objectives drive this initiative: preventing and monitoring strokes, applying artificial intelligence in personalized medicine, and developing digital health platforms specifically tailored for hospital environments.

The development of this technology began within the framework of the European research project AI-SPRINT. This technology leverages electrocardiogram (ECG) data collected from wearable devices such as smartwatches and integrates lifestyle information through a mobile app. In addition to this information, which addresses the first two types of stroke risk factors, Innostroke will incorporate a third risk dimension: genetic and molecular biomarkers.

“By using advanced artificial intelligence technology and high-performance computing (HPC), we aim to provide continuous, more precise risk monitoring that could represent a major advance in stroke prevention,” states Daniele Lezzi, BSC researcher in the Computer Sciences department and developer of the technology.

This progress could not only enable effective stroke risk monitoring but may also enhance prediction and prevention of neurovascular diseases through AI-based solutions and multi-omics data. Such data is an integrated set of biological data from multiple levels of molecular analysis within an organism, giving scientists a detailed, comprehensive view of an organism’s biological functioning.

“In collaboration with Israel Fernandez, principal investigator of the pharmacogenomics and stroke genetics group at the Sant Pau Hospital Research Institute, we seek to refine the predictive capabilities of the technology by incorporating genetic and molecular biomarkers. This multifaceted approach allows us to provide proactive, personalized medical care, improving patients' quality of life,” explains Davide Cirillo, BSC researcher in the Life Sciences department and an expert in personalized medicine.

Thanks to new funding from the Plan de Recuperación, Transformación y Resiliencia under the National Strategy for Artificial Intelligence, Innostroke will not only deliver numerous innovations, but will also seek to reach out to the market, to users and healthcare professionals to maximise its impact on stroke prevention. Additionally, the potential to apply this technology to similar conditions and/or other neuro and cardiovascular diseases will be explored, encouraging technological innovation generated by BSC.