The Innostroke Project Launches a Pilot Study to Advance Stroke Prediction

The Barcelona Supercomputing Center – Centro Nacional de Supercomputación (BSC-CNS), as part of the Innostroke project, has promoted a pilot study at Hospital de Sant Pau aimed at collecting data and methodologically validating personalized stroke risk prediction models in a real-world clinical setting.

The study, developed in collaboration with researchers from the Pharmacogenomics and Neurovascular Genetics group at the Institut de Recerca Sant Pau (IR Sant Pau) and the Neurology and Cardiology departments at Hospital de Sant Pau, will include the participation of people who have had a stroke or who are at high risk of having one.

During the study, participants will use smartwatches for three months, while researchers will have access to a digital platform that will integrate the data collected by these devices, providing an estimate of stroke risk together with a breakdown of all the indicators that have been analyzed. In the future, this tool could allow health care professionals to analyze information more comprehensively and support clinical decision-making.

One of the solution’s distinguishing features is its ability to interpret electrophysiological patterns derived from electrocardiograms (ECGs) as indirect markers of multiomic indicators, which are more costly to obtain and less accessible in routine clinical practice.

Through the use of artificial intelligence (AI) and high-performance computing (HPC), the system overcomes one of the main limitations of current solutions, which usually analyze data in isolation. Thanks to this multimodal integration, the study aims to make it possible to detect early risk signals through continuous monitoring over time.

“Thanks to the use of artificial intelligence and high-performance computing, we can train models capable of capturing complex relationships between clinical signals and biomarkers, advancing toward more robust and personalized prediction systems,” says Daniele Lezzi, principal investigator at the BSC and leader of the Innostroke project.

In addition to evaluating the platform’s performance, the pilot will make it possible to continue improving the performance of artificial intelligence models, as health care professionals will be able to access the system’s predictions and assess their usefulness and clinical consistency. This will help improve their accuracy and reliability, bringing these types of solutions closer to future implementation in clinical practice.

“This pilot gives us the opportunity to evaluate the technology in a real-world clinical setting and better understand how it can be integrated into care practice. Tools like this can help us identify at-risk patients earlier and move toward more personalized and preventive care,” says Cristina Gallego, researcher in the Pharmacogenomics and Neurovascular Genetics group at IR Sant Pau.

This pilot represents a key step for Innostroke, bringing technological innovation closer to the real-world clinical setting and laying the groundwork for more preventive, personalized and data-driven medicine.