Digital twin of the heart helps detect cardiac arrhythmias

Millions of people around the world live with cardiac arrhythmias, yet detecting and treating them accurately continues to be one of the great challenges of cardiovascular medicine. Now, a team led by the Universitat Politècnica de València (UPV) has launched a new non-invasive method to locate the origin of premature ventricular contractions (PVC) — one of the most frequent cardiac arrhythmias.

The method combines electrocardiographic imaging (ECGI) with digital twins of the heart, which helps to improve the accuracy with which the ‘focus’ where these arrhythmias originate can be identified. It has been published in the journal Computers in Biology and Medicine and was recently presented at EHRA 2025.

Currently, conventional electrocardiography (ECG) is used to detect premature ventricular contractions (PVCs). However, its results can be limited by anatomical differences between patients.

“Electrocardiographic imaging offers a more detailed view, but it also has certain limitations in accurately determining the exact point of origin of the arrhythmia,” said Andreu Climent, from UPV. “This is what we achieve with our new method.”

The system integrates ECGI with personalised heart simulations. The result is a “digital twin” capable of reproducing the electrical activity of the myocardium more faithfully, according to María S Guillem from UPV.

To create the new method, the research team generated a database of more than 600 simulations of cardiac arrhythmias using detailed anatomical models of the torso and the heart. Using these simulations, they developed an algorithm that allows the arrhythmia focus to be located with an average accuracy of 7.8 mm, compared to the more than 30 mm error recorded with standard ECGI, all in a completely non-invasive way.

In addition to the simulations, the new method was applied in a real clinical case of a patient with an arrhythmia localised in the free wall of the left ventricle. The model based on ECGI and digital twins achieved an estimation of the focus with an error of 15.5 mm, well below the error of 36.7 mm recorded with conventional ECGI.

“Our method can facilitate planning interventions, such as catheter ablation, by reducing the need for invasive scans and shortening intervention times,” said Jorge Sánchez from UPV. “It could be integrated as a support tool in the preoperative planning of ablations. And it would be especially useful in complex cases where other techniques are more limited, such as in arrhythmias originating in the septum or at the base of the ventricle.”

UPV’s COR-ITACA group is currently leading a new project called EPICA+, whose ultimate goal is to apply this method to many more arrhythmias and to apply it in clinical practice. To do this, they will integrate ECGI with artificial intelligence and digital heart twins.

“The combination of these three technologies will help to improve the diagnosis and treatment of cardiac arrhythmias,” said Guillem, project coordinator together with Climent.

The project brings together a multidisciplinary team of engineers, clinicians and data scientists, who will draw on a database of more than 1000 patients and 20,000 computational models. Clinical trials involving 144 patients will validate the technology with the aim of doubling treatment success rates and reducing costs.

Image credit: iStock.com/Prostock-Studio