AI brings healthcare to the edge of tomorrow
Artificial intelligence (AI) has the potential to have a major impact on the life sciences, especially healthcare. But delivering on that potential will require data – and lots of it. To get that data, developers must be willing to go to the edge. “AI’s benefits will largely come from using IT resources found at the periphery of the network, where the source of the data is produced,” explains Danilo Ardagna(opens in new window), an associate professor at the Polytechnic University of Milan(opens in new window). This is where edge computing comes into play. Unlike traditional cloud computing, where data is sent to a remote data centre for processing, edge computing is a distributed framework that brings processing power and data storage closer to the source of the data – typically near the end devices that generate and consume it. “The advantage of processing data locally is that it reduces latency, improves application performance and enables real-time insights,” says Ardagna.
Supporting the development of AI applications
Helping companies and researchers fully leverage these benefits is the EU-funded AI-SPRINT(opens in new window) project. “Our goal is to democratise access to AI technologies by simplifying and accelerating the development of artificial intelligence applications through edge computing,” adds Ardagna, who coordinated the project. At the heart of the project is a suite of novel tools designed to support the development of AI applications, along with securing their execution, ensuring easy deployment and optimising runtime management. “These tools are unique in that they allow for the performance, energy efficiency and accuracy of the AI models while simultaneously providing security and privacy guarantees,” notes Ardagna.
AI tools in the life sciences
Beyond designing the AI tools, the project also applied these tools to several use cases – including in the life sciences. For example, the Life Sciences Department of the Barcelona Supercomputing Center(opens in new window) (BSC) integrated AI-SPRINT’s technology to craft AI models for stroke risk assessment using sensor data and lifestyle information. Specifically, BSC used AI-SPRINT solutions to safeguard sensitive healthcare data and optimise its risk forecasting models. “Our unique framework not only enables the incorporation of wearable and mobile devices into healthcare but ensures that these devices can leverage AI to provide valuable insights into patient care,” remarks Ardagna. The BSC personalised healthcare use case was so successful that a spin-off company(opens in new window) was established to promote the use of its stroke risk assessment application. The company has since raised nearly EUR 700 000 in funding. BSC also partnered with the Sant Pau Stroke Research Centre in Barcelona to develop a more advanced solution for risk prevention that includes multi-omics data.
AI and edge computing revolutionising healthcare
An unexpected outcome of the project was BSC’s use of the AI-SPRINT models to acquire electrocardiogram (ECG) data from patients participating in a study conducted by the Ospedale Maggiore di Niguarda in Milan(opens in new window). Following the completion of the study, the participating clinicians identified a patient with significant cardiovascular and heart-related concerns. This timely intervention, made possible thanks to the use of AI models, allowed for the commencement of additional analyses and treatment, thereby mitigating potential health complications. “This is but one example of how artificial intelligence and edge computing will revolutionise healthcare research and patient care – and AI-SPRINT is proud to have contributed some of the tools that are making this revolution possible,” concludes Ardagna.
