Covid-19 comes with a range of symptoms – from a sore throat and the loss of taste to more serious ones like lung failure. But how can doctors predict how serious the disease will be when it first manifests? “The interaction between the viral infection, the host’s response, and the development of cardiovascular inflammation and injury is still poorly understood. It’s hard to know whether a patient’s symptoms will remain mild or rapidly deteriorate and trigger multiple organ failure,” says Adrian Ionescu, a professor at EPFL’s Nanoelectronic Devices Laboratory, within the School of Engineering. If doctors could use a scientific method to better understand and predict the likelihood of a patient’s condition worsening, they could more effectively screen patients at hospitals and deliver personalized care.
To that end, EPFL has launched a pan-European research program called Digipredict. The goal is to develop a digital twin that can detect serious complications in Covid-19 patients, employing breakthrough technology in the fields of artificial intelligence, smart patches and organs-on-chips. The initiative brings together around a dozen partner organizations (universities, hospitals and startups), including four EPFL labs: Ionescu’s along with the Embedded Systems Laboratory headed by David Atienza; the Laboratory of Movement Analysis and Measurement headed by Kamiar Aminian; and the Machine Learning and Optimization Laboratory headed by Martin Jaggi. Beyond contributing to the fight against Covid-19, the researchers hope that their technology will revolutionize the detection, monitoring and personalized treatment of inflammatory disease in general.
A “cytokine storm”
“We know that an excessive immune-system response can lead to serious cardiovascular dysfunction,” says Wolf Hautz, a professor at Inselspital university hospital in Bern and a member of the Digipredict team. “When the body detects a viral infection, it begins producing large amounts of cytokines, which are proteins that signal the immune response. But in some Covid-19 patients we’re seeing a ‘cytokine storm,’ which can seriously damage their cardiovascular systems. Being able to detect the first signs of these storms and track them in real-time would be a major step forward in the treatment of high-risk patients.”
Predicting disease progression, minute-by-minute
The Digipredict digital twin will be designed specifically for healthcare applications. It will consist of a smart patch with integrated technology for collecting a range of medical data, such as blood oxygen levels, breathing rate and body temperature. The patches will also include nanosensors linked to an artificial-intelligence program in order to continually track specific biomarkers that indicate a cytokine storm may be brewing. These biomarkers, located in a patient’s interstitial fluid, give an indication of the trajectory that the disease will follow. “Our digital twin will use organ-on-chip technology to select the right biomarker combination for generating an accurate picture of how the disease is progressing in a patient and how well the chosen treatments are working,” says Albert van den Berg, a professor at the University of Twente and another member of the Digipredict team. Ionescu adds: “With our data collection devices and AI algorithms, we can give doctors objective, quantitative information for making clinical decisions with as little error as possible.” The digital twin will be designed to follow minute-by-minute snapshots of patients’ conditions, helping doctors select personalized treatment protocols.
A prototype within two years
“This multi- and cross-disciplinary project will combine scientific excellence with engineering know-how, and leverage the expertise of doctors, biologists, electrical engineers, computer scientists, signal-processing engineers and social scientists from across Europe,” says Ionescu. In addition to EPFL, the other partner organizations are: the University of Twente, ETH Zurich, IMEC in Belgium, Stichting Imec in the Netherlands, the Charité university hospital in Berlin, the University of Bern (through Inselspital), and three startups (Ascilion, EPOS-IASIS and SCIPROM). These organizations have agreed to pool their knowledge to develop the first digital twin for Covid-19 patients. It will be tested at the two university hospitals involved in the program.
“This project combines the latest advancements in artificial intelligence and biomedical research,” says Alexander Meyer Chief Medical Information Officer at German Heart Center Berlin and professor of Clinical AI and Data Science at Charité. “Digipredict will change how doctors see and evaluate the trajectories of individual Covid-19 patients. And we hope that it will also improve the prevention, diagnosis, monitoring and treatment of cardiovascular disease.” Digipredict has received €6 million in funding over four years and will bring together around 50 scientists to develop the first device of its kind. A prototype should be ready in around two years; clinical trials will follow.