Image showing heart failure
Credit: Nikita John Creagh / Getty Images

Researchers from King’s College London, Imperial College London, and the Alan Turing Institute have created over 3,800 anatomically accurate digital replicas of human hearts. In a proof-of-concept study published in Nature Cardiovascular Research, the research team was able to produce personalized cardiac “digital twins” at an unprecedented scale using real patient data and clinical measurements. 

Digital twin technology enables the simulation of organs or tissues, providing insights into their physiology and function that may otherwise be challenging to measure directly through traditional experimental methods. In the current study, the cardiac digital twins were used to investigate how age, sex, and lifestyle factors influence the heart’s electrical function. 

The findings showed that age and obesity can change the heart’s electrical properties, which could provide an explanation for why these factors are linked to a higher risk of heart disease. Conversely, differences found in clinical readings between men and women were mainly attributed to different heart sizes rather than changes in electrical conductivity. 

“These insights will help refine treatments and identify new drug targets,” said Pablo Lamata, PhD, professor of biomedical engineering at King’s College London. “By developing this technology at scale, this research paves the way for their use in large population studies. This could lead to personalized treatments and better prevention strategies, ultimately transforming how we understand and treat heart diseases.”

Previous studies have reported on the potential of cardiac digital twins to support clinical decision-making and the selection of personalized treatment for cardiovascular conditions such as atrial fibrillation and cardiomyopathy. However, creating these digital replicas can be complex and time-consuming, which has so far limited the scale at which digital twin models can be built. 

Thanks to recent advances in statistical and artificial intelligence (AI) tools, the researchers were able to automate and accelerate the process of making cardiac digital twins. Using magnetic resonance imaging (MRI) and electrocardiogram (ECG) data from 3,461 patients registered in the UK Biobank, they developed a fully automated workflow that can create cardiac digital twins at an unprecedented speed and scale. These models were then validated in a separate cohort consisting of 359 patients with ischemic heart disease, which showed consistent changes in cardiac features relating to the patients’ age, sex, and body mass index (BMI). 

“Our research shows that the potential of cardiac digital twins goes beyond diagnostics,” said Steven Niederer, PhD, professor and chair in biomedical engineering at Imperial College London and senior author of the study. “By replicating the hearts of people across the population, we have shown that digital twins can offer us deeper insights into the people at risk of heart disease. It also shows how lifestyle and gender can affect heart function.”

Going forward, the study authors see great potential in using digital twin technology at a population scale to identify novel biomarkers and therapeutic targets, as well as to predict the therapeutic effects and potential side effects of both existing and emerging treatments. 

“The digital heart models we’ve built lay the foundation for the next step in our research—linking heart function to our genes,” said Shuang Qian, PhD, first author of the study and visiting research associate at King’s College London. “This could help us understand how genetic variations influence heart function in a way that’s never been done before [and] lead to more precise and personalised care for patients in the future.”

Also of Interest